Sorting cereal seeds

Abstract

A method of sorting seeds is disclosed. The seeds contain non-colored seeds and colored seeds or darker colored and lighter colored seeds. The method comprises: i. supplying a seed stream (120) to a sorting station (122), the sorting station (122) comprising at least one backlight device (130) for backlighting a seed of the seed stream (120) and at least one camera (132) for taking at least one image of the backlighted seed; ii. taking, with the camera (132), at least one image of the backlighted seed of the seed stream (120); iii. automatically identifying, from the image taken in step ii., seeds to be sorted out from the seed stream (120); and iv. automatically ejecting seeds identified to be sorted out from the seed stream (120), wherein the at least one backlight, in the HSL color space, has a H coordinate of H ≤ 150 or H ≥ 310 and a L coordinate of 0.10 ≤ L ≤ 0.80.

Description

[0001] SORTING CEREAL SEEDS

[0002] Technical Field

[0003] The present invention relates to a method of sorting seeds, to a sorting device for sorting seeds and uses of the sorting device. The method and the sorting device may specifically be used for sorting seeds containing a mixture of (darker) colored and (lighter or) non-colored seeds, e.g. as cereal seeds with a non-colored aleurone and cereal seeds containing a blue aleurone in the field of plant breeding. However, other fields of application comprising the sorting of seeds are also feasible.

[0004] Background art

[0005] In the field of plant breeding, such as such as cereal plant breeding, e.g. breeding of wheat, blue aleurone loci and / or genes may be useful color markers for some other genes and / or loci of interest being located on the same chromosome or even on the same chromosome arm as the color marker. Thus, the presence of a blue color in the aleurone of the seed may indicate the presence of the gene of interest. In general, the closer both genes and / or loci, the more reliable the presence of the gene of interest is indicated by the blue color as a higher distance generally increases occurrence of recombination and / or breakage. As an example, seeds may contain a recessive male sterility gene and / or a locus causing male sterility in plants lacking a restorer gene, and the gene of interest is a male fertility restorer gene linked to the blue aleurone color locus. The blue color may be used to distinguish fertile “blue” seeds containing the blue aleurone locus and indicating the presence of the restorer gene from non-colored ’’white” sterile seeds lacking the blue aleurone locus, indicating the absence of the restorer gene. Thus, these seeds containing a male sterility locus and / or gene and containing a fertility restorer gene linked to a blue aleurone locus can be used as maintainer seeds to reproduce male-sterile “white” seeds and fertile “blue” seeds as their selfed progeny contains both types of seeds. The sterile seeds may grow into useful female lines for hybrid seed production as they can only set seed after cross-pollination by a (male) fertile plant. As another example, the blue aleurone locus may be used in cereal seeds, such as wheat seeds, to identify higher amounts of anthocyanins in the seed, and / or to create new food products that are naturally colored rendering synthetic color product, e.g. in breakfast cereals, superfluous. The colored “blue” seeds may thus need to be sorted from non-colored “white” seeds in any production field or seed lot containing blue aleurone-containing seeds and “white” seeds.

[0006] WO 2023 / 088892 A1 describes a method for categorizing / sorting seeds, the method comprising the steps of: providing a sample including at least one seed; obtaining a near infrared, NIR, spectrum of at least a subset of the sample; determining presence of an organic colorant in at least the subset of the sample based on the obtained NIR spectrum; and categorizing / sorting at least the subset of the sample based on the determination. Based on this, mis-colored white seed and blue seed with a fading blue color can be categorized. Further, it is also said to be possible to distinguish between single blue and double blue seed. WO 2014 / 109993 A2 describes a system and method for separating seed or grain based on optical differences in the starch composition. The method for separating seed or grain based on optical differences in the starch composition includes receiving a seed group comprising a plurality of seeds. The method further includes illuminating each seed of the seed group from an illumination source disposed behind the seed such that the seed is back-illuminated. The method further includes sorting each seed of the seed group based on the differences in the starch composition. In some cases, the method includes sorting each seed by separating the seed group into the following groups: waxy seeds and non-waxy seeds.

[0007] Despite the advantages achieved by known methods and devices, several technical challenges remain. In general, standard separation machines separating colored seeds based on size and / or density may not work effectively since the size and / or density of colored seeds are the same as normal seed. Thus, separation by color may be necessary. However, color separators in existing seed processing equipment being used to separate colored seed from normal seeds may often provide low quality of sorting. The color separator may use an electronic eye for recognizing different colors. Seeds may pass the electronic eye and, in case a seed having a different color than the desired seed is identified, a sudden burst of air may be activated to push that seed into a reject bin while the rest of the seeds may pass to another bin. Thus, there is still a need for improvement of the sorting efficiency of optical separators, specifically with respect to sorting purity and / or sorting yield. In particular, there is still a need for sorting seeds carrying only one dose (1 n) of the BLA locus generally showing only weak coloration, which frequently leads to insufficiently sorting of seeds.

[0008] Problem to be solved

[0009] It is therefore desirable to provide devices and methods at least partially addressing above- mentioned technical challenges of known devices and methods. Specifically, it is an object of the present invention to provide devices and methods, which improve sorting purity and sorting yield in sorting of seeds containing a mixture of seeds having a normal (non-colored) or lightcolored seed and (darker) colored seed, such as cereal seeds having a normal (non-colored) aleurone and cereal seeds containing a blue aleurone, or disease-infected seeds, such as cereal seeds, with another seed color compared to non-disease-infected seeds, such as cereal seeds (such as darker seed color, a reddish seed color, brown / black / grey spots, etc.), or sorting darker colored seeds (e.g., containing more anthocyanins) from lighter colored seeds (to get more uniformly colored seeds).

[0010] Also, in the above-described hybrid system in cereals wherein blue aleurone color may be used to distinguish fertile “blue” seeds containing the blue aleurone locus (indicating the presence of the restorer gene) from non-colored ’’white” sterile seeds lacking the blue aleurone locus (indicating the absence of the restorer gene), it should be noted that the aleurone layer is triploid (3n). As there may be male transmission of the blue aleurone locus, seeds may contain one (1 n), two (2n) or three (3n) copies of the blue aleurone (BLA or BA) locus in the aleurone layer that will respectively result in light blue (1 n or 2n) seeds, and dark blue (3n) seeds. Also, a repeated amplification of a blue seed batch containing 3n blue aleurone seeds (also named diso- mic or double blue seeds) leads to significant reduced proportions of sterile (“white”) and maintainer (1n or 2n blue) seeds (since 3n BLA seeds only produce (3n) blue seed progeny) and hence to a reduced production of white seeds and maintainer seeds per area. On average, across different varieties, the proportion of white seeds in a harvested seed lot drops from approx. 64% to 25% over 3 amplification rounds. To keep the production of useful seeds for hybrid production and maintenance most effective, the amount of dark blue 3n seeds should be reduced by means of a specific seed sorting step in which either the darkest seeds of a blue seed fraction are shot out, or alternatively only light blue seeds are shot out and only the light blue fraction is used for seed amplification. Said sorting step can also be used to remove (darker) colored seed from a seed batch containing (darker) colored and non-colored or light-colored seeds.

[0011] Summary

[0012] This problem is addressed by a method of sorting seeds and by a sorting device for sorting seeds with the features of the independent claims. Advantageous embodiments which might be realized in an isolated fashion or in any arbitrary combinations are listed in the dependent claims as well as throughout the specification.

[0013] A first aspect of the present invention is a method of sorting seeds, wherein the seeds contain darker colored and lighter colored seeds, or non-colored seeds and colored seeds, or the seeds can have been pre-sorted to remove colored from non-colored seeds so that the seeds at least contain light colored seeds and darker colored seeds, and may contain non-colored seeds.

[0014] The term “sorting” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process of differentiation and separation. The sorting may comprise at least two process steps, which, specifically, may be performed in parallel and / or in a timely overlapping fashion, wherein, in a first step, objects may be differentiated from each other according to one or more sorting criteria, wherein, in a second step, the objects may be separated from each other based on the differentiation. Specifically, the differentiation of object may comprise assigning objects into at least two different sorting categories. The separation may be performed based on the sorting categories, wherein objects being assigned to a first sorting category may be physically separated from objects being assigned to a second sorting category. The sorting of seeds, such as cereal seeds, as described herein may be performed on any seed sorting device allowing adjustable background light color, e.g. on a ASM® EUREKA sorter, on a Buhler ® Sortex H sorter, a H series color sorter from AnySort (Anhui Jiexun Optoelectronic Technology Co., Ltd), and / or on 3U Vision ® OPTICA sort models or the like. The sorting may specifically comprise a sensor-based sorting. Specifically, the differentiation may comprise using one or more sensor devices for sensing the sorting criteria. The sorting may comprise an optical sorting. Thus, as an example, the differentiation of the objects to be sorted may be at least partially performed by optical means. The sorting may comprise differentiating the objects to be sorted based on a color and / or at least one color coordinate, as will be outlined in further detail below, and, thus, may comprise determining at least one color and / or at least one color coordinate of the objects and assigning the objects into at least two different sorting categories according to the determined color and / or color coordinate. The sorting may specifically comprise automatically sorting objects to be sorted, in particular without manual action and / or interaction with a user, such as by using a combination of one or more sensor devices for enabling differentiation and one or more controllers for enabling separation, such as by actuation of at least one separation device. A result of the sorting may comprise at least two distinct groups of objects, wherein each object in a specific group may have at least one common property or characteristic with the other objects of this specific group. Specifically, the sorting may result in a separation of the objects to be sorted into at least two distinct groups, wherein objects in a first group may have a specific color and / or at least one specific color coordinate, wherein objects in a second group may be different from the specific color and / or the at least one specific color coordinate. For example, the sorting may comprise differentiating non-colored cereal seeds as defined herein from colored cereal seeds, such as cereal seeds containing a blue aleurone and separating the non-colored cereal seeds from the colored cereal seeds, such as cereals seeds containing a blue aleurone, or differentiating dark blue seeds from light blue seeds (and any remaining non-colored seeds) or differentiating light blue seeds from dark blue seeds (and any remaining non-colored seeds).

[0015] The term “seed” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a unit of reproduction of a flowering plant, capable of developing into another such plant, i.e. by producing functional pollen or male gametes. As such, the term may specifically refer to a fertile seed. The term may also refer to a sterile seed. The term “sterile seed” in connection with the present invention may refer to seeds growing into plants failing or partially failing to produce functional pollen or male gametes (also known as male sterility). This can be due to natural or artificially introduced genetic predispositions or to human intervention on the plant in the field. Male sterility / fertility in cereals, such as wheat, can be reflected in seed set upon selfing, e.g. by bagging heads to induce self-fertilization. Likewise, fertility restoration can also be described in terms of seed set upon crossing a male sterile plant with a plant carrying a functional restorer gene, when compared to seed set resulting from crossing (or selfing) fully fertile plants. A male parent (or pollen parent), is a parent plant that provides the male gametes (pollen) for fertilization, while a female parent or seed parent is the plant that provides the female gametes for fertilization, said female plant being the one bearing the (hybrid) seeds. Male sterility can be restored, for example, by introducing a functional restorer gene into the genome of the sterile plant. The seeds to be sorted may comprise, specifically may be, cereal seeds, specifically lighter colored and / or non-colored cereal seeds and colored cereal seeds, more specifically lighter colored and / or non-colored cereal seeds and darker-colored cereal seeds.

[0016] The term “cereal seed” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a seed of cereal. Specifically, the cereal seed may comprise a grain of cereal. The term “grain”, as used herein, may be equivalent to seeds, and may include grains for sowing and / or planting a crop. The cereal seed may be a whole grain of cereal. The cereal seed may specifically be a grain of cereal comprising an endosperm, a germ, an aleurone layer, a seed coat and a pericarp. The cereal seed may be a grain of a cereal selected from the group consisting of: corn; rice; wheat; barley; sorghum; millet; oat; rye; triticale. The cereal seed may specifically be a grain of wheat.

[0017] “Wheat”, as used herein, may refer to plants from the genus Triticum, including but not limited to common / bread wheat Triticum aestivum or T. aestivum), emmer wheat ( T. dicoccum), einkorn wheat ( T. monococcum), durum wheat ( T. durum), khorasan wheat ( T. turanicurri), or spelt wheat ( T. speita), specifically hexapioid T. aestivum or T. speita, and tetrapioid T. durum, including wheat referred to as hard or soft wheat (based on endosperm texture), winter or spring wheat (based on sowing season), and red or white wheat (based on seed coat color).

[0018] The seeds may contain lighter colored seeds and darker colored seeds, or non-colored seeds and colored seeds.

[0019] The term “non-colored” as used herein specifically may refer, without limitation, to a natural phenotype of seeds, such as cereal seeds, that have a lighter color than the colored seeds that are to be shot out, or that lack a color / pigment present in the colored seeds that are to be shot-out, from a mix of colored and non-colored seeds, e.g., red or white wheat seeds that have no blue aleurone are non-colored seeds when the colored seeds have a blue aleurone. The non-colored seeds include cereal, such as wheat, seeds with a white or red seed coat lacking a blue aleurone and having no purple pericarp, when the colored seeds to be shot-out from a mixture of colored and non-colored seeds are wheat seeds with a blue aleurone or purple pericarp (e.g., examples of wheat varieties with white seed color are the varieties Blini (spring wheat type) and Heroldo (winter wheat type), and examples of wheat varieties with red / reddish seed color are the varieties Granary (spring wheat type) and Solehio (winter wheat type)). Non-colored seeds, as used herein, may refer to the majority of cereal crops grown commercially for producing food, feed or drinks, excluding small / niche market cereal seeds with purple or blue seed color. The non-colored seeds as used in the context of shooting out colored seeds having a blue aleurone from a mixture of said colored seeds and non-colored seeds may also be referred to as “white” seeds herein, which specifically also refers to known red or white seeds based on seed coat color (like hard red winter or soft white spring wheat), to distinguish from seeds with a blue aleurone or purple pericarp. The colored wheat seeds may specifically comprise red seeds, when to be sorted out from a mixture of colored and non-colored wheat seeds, wherein the non-colored wheat seeds have a white seed color and the colored wheat seeds to be shot-out have a red seed color. As used herein, non-colored seeds may also refer to non-infected seeds, when sorting a mixture of disease-infected seeds and non-infected seeds, wherein the disease-infected seeds have a darker seed color or have darker spots when compared to the non-infected seeds.

[0020] The term “colored” as used herein specifically may refer, without limitation, to seeds having a darker color than the non-colored or normal colored seeds, such as having darker seed color than the non-colored seeds or darker spots on the seed. For example, the colored seeds may comprise seeds having a blue aleurone and lacking a purple pericarp color. Alternatively or additionally, the colored seeds may comprise disease-infected seeds, such as seeds having a fungal, bacterial or viral infection that changes seed color to a darker colored (seen as darker (dark brown or black or grey) spots on the seed, overall darker seeds, or a change in color from white to red / pink seed color, such as infection by ergot (Ciaviceps, such as C. purpurea), bunt, smut, smudge, black point, etc., including any secondary disease infections). In one embodiment, dis- ease-infected seeds, such as cereal seeds, may refer to seeds with dark brown or black spots on the seeds, such as cereal seeds. Whether seed is colored or darker colored compared to a (non-colored or lighter colored) control seed, as used herein, can be seen by eye or can be measured by videometer, a colorimeter or spectrophotometer (e.g. by determination of the lightness). In one embodiment of the invention, the sorting methods as described herein can be used as a quality control tool to quantify the amount of disease-infected (such as ergot-infected) or darker-colored seeds, such as cereal seeds, in a seed batch, which can determine if the seed batch meets the requirements of a certain market / channel, such as if it is acceptable for release as food or feed product, or needs to be destroyed or directed to other (such as non-food / non- feed) markets / channels. The colored seeds may specifically be seeds having a natural seed coat color and a blue aleurone and normal pericarp color, such as wheat seeds with a white or red seed coat color having a blue aleurone and having no purple pericarp. A blue aleurone color may be associated with the presence of anthocyanins in the aleurone layer of cereal seeds. The blue color of seeds may be genetically controlled by a blue aleurone locus, which is involved in the biosynthesis of anthocyanin. Current standard commercial wheat seeds may not contain a blue aleurone, but the trait can be introduced in wheat by introgressing it from some species of Triticeae, or by adding a blue aleurone locus to plants lacking it by plant transformation or (targeted) genome editing. Several blue aleurone loci conferring the blue aleurone trait may have been transferred into wheat from Triticeae species plants such as Thinopyrum ponticum, Ag- ropyron eiongatum, Triticum boeoticum, Triticum monococcum, or Thinopyrum bessarabicum. The seeds containing a blue aleurone may also be referred to as “blue” seeds (or BLA seeds) herein. The colored seeds may specifically refer to both light colored seeds, such as seeds having a 1n or a 2n BLA locus in the aleurone layer, and dark colored seeds, such as seeds having a 3n BLA locus in the aleurone layer, as will be outlined in further detail below. In one embodiment, the colored cereal seeds as used herein may also have a purple pericarp and a normal (not blue) aleurone and the non-colored cereal seeds as used herein may have a natural seed color such as white or red seed color (with a normal (not a purple) pericarp and aleurone (not blue) color). The colored cereal seeds may also have a darker seed color, such as dark red rice seeds, and the non-colored cereal seeds may have a lighter seed color, such as light red rice seeds, wherein the dark seeds can be separated from the lighter seeds to get a more uniform seed batch. Alternatively, the colored seeds may be a sub-fraction of seeds which contains re- mainings of an undesired part of the seed in processing, such as remainings of the aleurone layer in polished seeds (e.g. rice) or the remainings of the testa from peeled seeds (e.g. peanuts).

[0021] The term “blue aleurone locus” (or BLA locus) as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to the genetic locus causing a blue aleurone phenotype, as can be transferred by introgression in wheat from related species. The seeds containing a blue aleurone are seeds comprising a BLA locus, specifically wheat seeds comprising a BLA locus, and may be referred to as blue seeds.

[0022] For example, the cereal seed may comprise wheat seeds. A 2-line male sterility system may be used with a maintainer plant producing white sterile and blue fertile cereal seeds upon selfing. Any regular wheat plant may act as male parent to restore fertility. The BLA locus and a restorer gene may be located on a monosomic addition chromosome (42+1 Chr plants) or on a homoe- ologous chromosome pair (42 Chr plants). The BLA locus and the restorer gene may be either on the same or different arms of the same chromosome. The aleurone layer may be triploid, such as be being part of the endosperm that is triploid (3n), and as there may be male transmission of the BLA locus, seeds may contain 1 (1 n), 2 (2n) or 3 (3n) copies of the BLA locus in the aleurone layer that will respectively result in light blue seeds, blue seeds, or dark blue seeds. For possible embodiments of such a 2-line male sterility system, reference may be made to e.g. Whitford et al., 2013, J. Exp. Botany 64 (18): 5411-5428, and Zhou et al., 2006, CropScience 46:250-255, CN 100420368, WO 2019 / 043082 A1 , WO 2020 / 056259 A1 and WO 2023 / 005883 A1 . The BLA locus and the restorer gene may preferably be located on the same chromosome arm of a monosomic addition chromosome, or on the same arm of one of the 2 chromosomes in a homoeologous chromosome pair. Thus, these genes may be closely linked, or alternatively the BLA locus and the restorer gene may be each located on another chromosome arm of the same addition or homoeologous chromosome. As an example, the restorer gene may be a MS1 , MS5, MS9, MS22, MS26, or MS45, specifically depending on what causes the male sterility. For example if a mutation or inactivation or deletion of an MS1 gene causes male sterility, then MS1 may be the restorer gene to use, and if mutation or inactivation or deletion of all MS45 genes (on each wheat sub-genome (A, B and D)) causes male sterility, then MS45 may be the restorer gene to use. Alternatively or additionally, the blue aleurone genes / loci may be one of the genes / loci as described in US 11 ,390,877 B2, WO 2019 / 043082 A1 or WO 2020 / 056259 A1. The blue aleurone locus may be obtainable or obtained, e.g., from Agropyron elongatum, Agropyron trichophorum, Triticum boeoticum, Triticum monococcum, Triticum thaou- dar, Triticum aestivum, or Thinopyrum ponticum or from wheat lines having an introgressed BLA locus or may be from known seed accessions Sebesta Blue, Blue Sando, Blue Baart, Blue Onas, Blue 1 , PBB, or Blue Norco. In one embodiment of the invention, the male-sterile female plants of the invention may comprise triple homozygous mutations of the MS45 male-fertility polynucleotide in wheat, which mutations cause a male sterility phenotype. In one embodiment, such male-sterile plants can be obtained from a 2-line hybrid system in wheat that comprises triple homozygous mutations of a MS45 male-fertility polynucleotide (such as EMS mutations inactivating each MS45-A, MS45-B and MS45-D gene, or mutations inactivating each MS45-A, MS45-B and MS45-D gene as obtained by genome editing methods), and a plant restoration donor chromosomal component comprising a 4E chromosomal component from Thinopyrum or Agropyron, the 4E chromosomal component comprising: a) a plant polynucleotide that confers a plant seed phenotype (such as seed color, e.g. blue aleurone, P gene, anthocyanin, or Kala 4); and (b) a MS45 male-fertility restoration locus restoring male-fertility in a ms45 male-sterile wheat plant by the 4E chromosomal component, wherein expression of the 4E chromosomal component functionally complements the male-sterility phenotype from the triple homozygous MS45 mutations so that the wheat plant is male-fertile. In one embodiment, the plant polynucleotide that confers the plant phenotype (such as seed color) may be located on the same chromosomal arm of the 4E chromosomal component as the MS45 male-fertility restoration locus (not separated by a centromere). In one embodiment, said MS45 hybrid system and the male-sterile female wheat plant may be as described in W02020056259.

[0023] The colored seed may be a disease-infected seed with darker seed color or darker spots compared to the non-infected seed that is lighter or has no darker spots.

[0024] The method comprises the following steps that may be performed in the given order. However, a different order may also be possible. In particular, one, more than one or even all of the method steps may be performed once or repeatedly. Further, the method steps may be performed successively or, alternatively, one or more of the method steps may be performed in a timely overlapping fashion or even in a parallel fashion and / or in a combined fashion. The method may further comprise additional method steps that are not listed.

[0025] The method comprises: i. supplying a seed stream to a sorting station, the sorting station comprising at least one backlight device for backlighting a seed of the seed stream and at least one camera for taking at least one image of the backlighted seed;

[0026] II. taking, with the camera, at least one image of the backlighted seed of the seed stream; ill. automatically identifying, from the image taken in step ii . , seeds to be sorted out from the seed stream; and iv. automatically ejecting seeds identified to be sorted out from the seed stream, wherein the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80.

[0027] Specifically, the at least one backlight, in the HSL color space, may have a H coordinate of H < 150 or H > 310 and a L coordinate of 0.20 < L < 0.80. The term “supplying” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process of making available for further processing. The supplying may specifically comprise providing the seed stream to the sorting station.

[0028] The term “seed stream” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a plurality of seeds. The seed stream may comprise a plurality of seeds, wherein the plurality of seeds may be provided to the sorting station in a directional fashion. Additionally or alternatively, the seed stream may comprise a plurality of seeds arranged in a regular fashion. The seed stream may comprise a plurality of seeds, wherein the seeds of the seed stream may be arranged individually in a row or line. Thus, as an example, the seeds of the seed stream being supplied to the sorting station may comprise a plurality of seed being provided one by one to the sorting station.

[0029] The term “sorting station” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a device or combination of devices configured for performing at least one sorting function. Specifically, the sorting station may comprise at least one sensor device, such as the camera, configured for differentiation of the seeds in the seed stream. The sorting station may further comprise at least one ejector configured for separation of the seeds in the seed stream by ejecting seeds to be sorted out from the seed stream. The sensor device and the ejector of the sorting station may be connected with each other, e.g. via one more controller, such that, upon differentiation of seeds in the seed stream using the sensor device, the ejector ejects the respective seeds from the seed stream. The sorting station may further comprise at least one front light device for illuminating the seed stream on a front side. The sorting station may comprise two, four or even more front light devices. The front light device may comprise a white LED.

[0030] The term “backlight” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to light being provided from behind. The backlight specifically may be or comprise light present in the background. As used herein, the term “light” may refer, without limitation, to electromagnetic radiation in the visible light spectral range. Herein, the term “visible spectral range”, generally, may refer to a spectral range of 380 nm to 760 nm. The method may comprise using different backlights, such as different backlights in different, specifically repeated, sorting steps. In case the method may comprise more than on backlight, the backlight involved in the same sorting step may have constant light settings.

[0031] The term “backlight device” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a device configured for generating light in the sense of the above-mentioned definition. The backlight device may specifically comprise at least one light source for generating light. For example, the backlight device may comprise at least one light-emitting diode (LED). The backlighting device may specifically comprise multiple LEDs, such as at least one red LED, at least one green LED and at least one blue LED. The multiple LEDs may form a multicolor white LED configured for emitting light at least in the visible spectral range. The multicolor white LED may be configured for providing different colors in the visible spectral range according to a color mixing from the multiple LEDs. The color mixing in the multicolor white LED may be controllable by controlling the multiple LEDs individually. The backlight device may be arranged behind the seed stream in a field of view of the camera and, thus, may provide “backlight”.

[0032] The colored backlight may be provided directly by the backlight device, specifically by controlling multiple differently colored LEDs to achieve the respective color of the backlight, e.g. adjustable RGB LEDs. Alternatively or additionally, the colored backlight may be provided using monochrome light or white light in combination with one or more color filters to achieve the respective color of the backlight. Thus, in this example, the backlight device may additionally comprise the one or more color filters. Alternatively or additionally, a colored background may be used being illuminated by a monochrome or white color backlight device. For example, a colored background plate or sheet or conveyor belt may be arranged behind the seed stream so as to provide the respective color of the backlight.

[0033] The term “camera” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a device having at least one imaging element configured for recording or capturing spatially resolved one-dimensional, two-dimensional or even three-dimensional optical data or information. As an example, the camera may comprise at least one camera chip, such as at least one CCD chip and / or at least one CMOS chip configured for recording images. As used herein, without limitation, the term “image” specifically may relate to data recorded by using a camera, such as a plurality of electronic readings from the imaging device, such as the pixels of the camera chip.

[0034] The camera, besides the at least one camera chip or imaging chip, may comprise further elements, such as one or more optical elements, e.g. one or more lenses. As an example, the camera may be a fix-focus camera, having at least one lens which is fixedly adjusted with respect to the camera. Alternatively, however, the camera may also comprise one or more variable lenses which may be adjusted, automatically or manually.

[0035] The camera specifically may be a color camera. Thus, such as for each pixel, color information may be provided or generated, such as color coordinates for three colors, e.g. H (hue), S (saturation), L (lightness) and / or R, G, B. A larger number of color values is also feasible, such as four colors for each pixel, for example R, G, G, B. Color cameras are generally known to the skilled person. Thus, as an example, each pixel of the camera chip may have three or more different color sensors, such as color recording pixels like one pixel for red (R), one pixel for green (G) and one pixel for blue (B). For each of the pixels, such as for R, G, B, values may be recorded by the pixels, such as digital values in the range of 0 to 255, depending on the intensity of the respective color. Instead of using color triples such as H, S, L and / or R, G, B, as an example, quadruples may be used, such as R, G, G, B or C, M, Y, K or the like. The color sensitivities of the pixels may be generated by color filters or by appropriate intrinsic sensitivities of the sensor elements used in the camera pixels. These techniques are generally known to the skilled person.

[0036] The term “taking at least one image” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to one or more of imaging, image recording, image acquisition, image capturing. The term “taking at least one image” may comprise capturing a single image and / or a plurality of images such as a sequence of images. For example, the taking of the image may comprise recording continuously a sequence of images such as a video or a movie. The taking of the at least one image may be initiated automatically, e.g. once the presence of the seed stream within a field of view and / or within a predetermined sector of the field of view of the camera is automatically detected. These automatic image acquisition techniques are known e.g. in the field of automatic barcode readers, such as from automatic barcode reading apps. The taking of the images may take place, as an example, by acquiring a stream or “life stream” of images with the camera, wherein one or more of the images, specifically automatically, are stored and used as the at least one image of the backlighted seed. The image acquisition may be supported by a controller, such as by at least one processor of the controller, and a storing of the images for image processing and / or evaluation may take place in a data storage device of the controller.

[0037] As outlined above, from the image taken in step ii. , the seeds to be sorted out are automatically identified and the seeds identified to be sorted out from the seed stream are automatically ejected. The term “automatically” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process which is performed completely by means of at least one computing unit, in particular without manual action and / or interaction with a user. The term “automatically” may specifically refer to any process which is performed by means of a controller.

[0038] The term “identifying” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process of recognizing one or more features in an image. The identifying may specifically comprise at least one image processing step for determining seeds to be sorted out in the image. For example, the identifying may comprise at least one image processing step for identifying seeds in the image. The identifying may further comprise at least one image processing step for determining if the seeds identified in the image are seeds to be sorted out. Optionally, the identifying may further comprise at least one object detection step, specifically at least one moving object detection step, such as for estimating a motion of the seeds in the seed stream, e.g. for estimating a motion of the seeds to be sorted out in the seed stream.

[0039] The term “seed to be sorted out” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a seed of the seed stream which is to be separated from the other seeds of the seed stream. The seed to be sorted out may be defined previously to be separated from the other seeds of the seed stream, e.g. by defining a specific phenotype of a seed which is to be separated from other seeds of the seed stream differing from the specific phenotype. For example, the seeds to be sorted out from the seed stream may specifically comprise the (darker) colored seeds, such as seeds containing a blue aleurone.

[0040] The term “ejecting” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process of removing an object from a group of objects. Specifically, the ejecting may comprise removing the seeds to be sorted out from the seed stream. The ejecting may comprise removing the seeds to be sorted out by means of at least one of a mechanical device and a pneumatic device. For example, a pneumatic ejector may be configured for ejecting the seed to be sorted out from the seed stream by using compressed air, such as by using compressed air directed via nozzles to separate the seed to be sorted out from the seed stream. By ejecting the seeds to be sorted out from the seed stream, the seeds in the seed stream may be separated into a first group of seed comprising the seeds to be sorted out and a second group of seeds comprising the non-ejected seeds of the seed stream.

[0041] As outlined above, the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The term “color space” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to an arbitrary coordinate system by which a color of an object, such as a color of a seed or a color of the backlight, may be characterized, such as mathematically or physically. Various color coordinate systems are generally known to the skilled person, such as color coordinate systems defined by the CIE (Commission internationale de I'eclairage). Color coordinate systems other than those defined by the CIE are also feasible. The color coordinates, in their entirety, may span or define the color space, such as by defining three or four basis vectors. Thus, when the camera captures an image of an object, a value for each color coordinate is generated by the camera for each pixel. As an example, the camera chip may contain color sensors recording values for each color, such as triples like HSL and / or RGB (Red Green Blue) and / or L*a*b or quadruples like CMYK (cyan, magenta, yellow, key), wherein the values are dependent on the sensitivity of the camera chip. The term “color coordinate” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to the coordinate of an arbitrary color coordinate system used for describing a color using coordinates. Several color coordinate systems are generally known to the skilled person and may also be used in the context of the present invention. Thus, as an example, a colorimetric coordinate system or a coordinate system may be used which is based on the human perception, such as the CIE 1964 color space, the M unsell color system or other coordinate systems, such as H, S, L and / or R, G, B and / or L, a, b.

[0042] The term “HSL color space” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a cylindrical-coordinate representation of points in an RGB color model. Specifically, the HSL color space may be defined by the color coordinates hue (H), saturation (S) and lightness (L). As used herein, the term “hue” may refer, without limitation to a color coordinate describing an angle around a central vertical axis of the color cylinder of the HSL color space. The hue may specifically be an angle from 0 to 360, wherein H =0=360 defines a red color, wherein H=30 defines an orange color, wherein H=60 defines a yellow color, wherein H=90 defines a yellow-green color, wherein H=120 defines a green color, wherein H=150 defines a green-cyan color, wherein H=180 defines a cyan color, wherein H=210 defines a cyan-blue color, wherein H=240 defines a blue color, wherein H=270 defines blue-magenta color, wherein H=300 defines a magenta color and wherein H=330 defines a magenta-red color. As used herein, the term “saturation” may refer, without limitation to a color coordinate describing a distance from a central vertical axis of the color cylinder of the HSL color space. The saturation may be defined in relative terms, specifically from 0 to 1 , and / or in absolute terms on a scale from 0 to 255, wherein full saturation in relative terms of 1 corresponds to an absolute value of 255. The saturation may describe a ratio of colorfulness to brightness of a color. Full saturation may be described by the value 1 , or alternatively by the value 255, and zero saturation may be described by the value 0. The saturation may comprise a percentage from 0% to 100%, equivalent to a scale from 0 to 1 , indicating a balance of a pure color and white. For example, with a hue of pure green (H=120), a saturation of 100% may be composed of only green light, specifically without red or blue light. Thus, changing the saturation on a primary color up and down the percentage range may be equivalent to increasing or decreasing the other two primary colors in equal amounts. A hue of pure green (H=120) with a saturation of 50% may indicate that the red and blue colors are increased by 50% of their value each. As used herein, the term “lightness” may refer, without limitation to a color coordinate describing a distance along a central vertical axis of the color cylinder of the HSL color space. The lightness may be defined in relative terms, specifically from 0 to 1 , and / or in absolute terms on a scale from 0 to 255, wherein full lightness in relative terms of 1 correspond to an absolute value of 255. The lightness may describe a brightness of a color relative to the brightness of a similarly illuminated white. A full lightness value of 1 may correspond to white, wherein a lightness value of 0 may correspond to black. Unless indicated to the contrary, the HSL values herein may be described on a hue scale with 0 < H < 360, on a relative lightness scale with 0 < L < 1 and on a relative saturation scale with 0 < S < 1 . Thus, S and L values given having a value in the range of 0 to 1 or given in % are always on the relative scale, whereas other values, specifically integer values above 1 , are always on the absolute scale from 0 to 255 for S and L or from 0 to 260 for H.

[0043] Without narrowing the scope of the invention, the invention will be specifically described with respect to the HSL color space. It shall be noted, however, that using other color spaces, such as those named above as well as further color spaces, is also feasible. For example, colors in the

[0044] RGB color space, wherein R, G, B e [0, 1], may be transformed into colors in the HSL color space according to the following equations: 2 wherein M = max(R, G, B) and m = min(R, G, Bf wherein S, L e [0, 1] in equations (2) and (3).

[0045] Similarly, colors in the HSL color space, wherein He [0, 360) and S, L e [0, 1], may be transformed into colors in the RGB color space according to the following equations: C = (1 - |2 * L - 1|) * S (4),

[0046] (R, G, B) = (R + m, G±+ m, B±+ m) (9).

[0047] Thus, any color or color coordinate given in the following in the HSL color space may be equally described by a color or color coordinate in the RGB color space and / or any other color space, or vice versa. For example, color converters on the World Wide Web may also be used for converting color from one color space to another, such as convertacolor.com or, e.g. a color converter for RGB to HSL conversion on the world wide web at www.w3schools.com / colors / col- ors_hsl.asp, a color converter for RGB to HSL conversion on the world wide web at www.rapidtables.com / convert / color / rgb-to-hsl.html and / or a color converter for HSL to RGB conversion on the world wide web at www.rapidtables.com / convert / color / hsl-to-rgb.html.

[0048] The at least one backlight, in the HSL color space, may specifically have a H coordinate in at least one range selected from the group consisting of:

[0049] 0 < H < 25 or H > 310, specifically except a range of 332 < H < 338;

[0050] 70 < H < 150;

[0051] 25 < H < 70.

[0052] Specifically, the H coordinate in the range 0 < H < 25 or H > 310 may comprise red backlight. The H coordinate in the range 70 < H < 150 may comprise green backlight. The H coordinate in the range 25 < H < 70 may comprise yellow backlight.

[0053] The at least one backlight, in the HSL color space, may have a S coordinate of 0.25 < S < 1 .0, specifically of 0.5 < S < 1 .0, more specifically of 0.75 < S 1 .0.

[0054] The at least one backlight, in the HSL color space, may have a H coordinate in the range of 0 < H < 20, a S coordinate in the range of 0.3 < S < 1 .0 and a L coordinate in the range of 0.28 < L < 0.78. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 20 < H < 25, a S coordinate in the range of

[0055] 0.3 < S < 1.0 and a L coordinate in the range of 0.36 < L < 0.78. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 25 < H < 45, a S coordinate in the range of 0.3 < S < 1 .0 and a L coordinate in the range of 0.63 < L < 0.78. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 45 < H < 50, a S coordinate in the range of

[0056] 0.65 < S < 1.0 and a L coordinate in the range of 0.56 < L < 0.72. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 50 < H < 55, a S coordinate in the range of 0.6 < S < 1 .0 and a L coordinate in the range of 0.44 < L < 0.72. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 55 < H < 60, a S coordinate in the range of

[0057] 0.6 < S < 1 .0 and a L coordinate in the range of 0.35 < L < 0.56. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 60 < H < 70, a S coordinate in the range of 0.6 < S < 1 .0 and a L coordinate in the range of 0.35 < L < 0.63. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 70 < H < 85, a S coordinate in the range of

[0058] 0.6 < S < 1.0 and a L coordinate in the range of 0.44 < L < 0.50. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 85 < H < 100, a S coordinate in the range of 0.6 < S < 1 .0 and a L coordinate in the range of 0.28 < L < 0.63. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 100 < H < 105, a S coordinate in the range of 0.3 < S < 1 .0 and a L coordinate in the range of 0.28 < L < 0.55. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 105 < H < 125, a S coordinate in the range of 0.3 < S < 1.0 and a L coordinate in the range of 0.28 < L < 0.50. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 125 < H < 150, a S coordinate in the range of 0.3 < S < 1 .0 and a L coordinate in the range of 0.28 < L < 0.50. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 310 < H < 332, a S coordinate in the range of 0.5 < S < 1 .0 and a L coordinate in the range of 0.13 < L < 0.45. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 338 < H < 350, a S coordinate in the range of 0.8 < S < 1 .0 and a L coordinate in the range of 0.35 < L < 0.50. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 338 < H < 350, a S coordinate in the range of 0.3 < S < 1 .0 and a L coordinate in the range of 0.55 < L < 0.78. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate in the range of 350 < H < 360, a S coordinate in the range of 0.5 < S < 1 .0 and a L coordinate in the range of 0.28 < L < 0.78.

[0059] The at least one backlight, in the HSL color space, may have a S coordinate of 0.25 < S < 1 and a H coordinate of 25 < H < 70. Thus, for yellow backlight, the saturation of the backlight may be in the range of 0.25 < S < 1 . Further, the at least one backlight, in the HSL color space, may have a L coordinate of 0.35 < L < 0.78. The at least one backlight, in the HSL color space, may have a H coordinate of 45 < H < 50, a S coordinate of 0.65 < S < 1 and a L coordinate of 0.55 < L < 0.65. For example, the at least one backlight, in the HSL color space, may have a H coordinate of 47, a S coordinate of 1 .0 and a L coordinate of 0.59. This specific yellow color may correspond to a RGB color of (R,G,B)=(255 / 210 / 46). These specific yellow colors may achieve best purity and lower product yield for light seed lots and very good purity and good product yield for dark seed lots compared to the other backlight colors, as will be outlined in further detail below.

[0060] Additionally or alternatively, the at least one backlight, in the HSL color space, may have a S coordinate of 0.33 < S < 1 .0 and a H coordinate of 70 < H < 150. Thus, for green backlight, the saturation of the backlight may be in the range of 0.33 < S < 1 .0. Further, the at least one backlight, in the HSL color space, may have a L coordinate of 0.28 < L < 0.63. The at least one backlight, in the HSL color space, may have a H coordinate of 109 < H < 113, a S coordinate of 0.3 < S < 1 .0 and a L coordinate of 0.45 < L < 0.50. For example, the at least one backlight, in the HSL color space, may have a H coordinate of 111 , a S coordinate of 1 .0 and a L coordinate of 0.48. This specific green color may correspond to a RGB color of (R,G,B)=(37, 245,0). Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 101 , a S coordinate of 1 .0 and a L coordinate of 0.45. This specific green color may correspond to a RGB color of (R,G,B)=(73, 230,0). These specific green backlights may achieve very good purity and lower product yield compared to the other backlight colors, as will be outlined in further detail below. Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 69 < H < 73, a S coordinate of 0.6 < S < 1 .0 and a L coordinate of 0.45 < L < 0.50. For example, the at least one backlight, in the HSL color space, may have a H coordinate of 71 , a S coordinate of 1 .0 and a L coordinate of 0.48. This specific green color may correspond to a RGB color of (R,G,B)=(78, 245,0). These specific green backlights may achieve the best overall purity and low product yield compared to the other backlight colors, as will be outlined in further detail below.

[0061] Additionally or alternatively, the at least one backlight, in the HSL color space, may have a S coordinate of 0.40 < S < 1.0 and a H coordinate of 25 < H or H > 310. Thus, for red backlight, the saturation of the backlight may be in the range of 0.40 < S < 1 .0. Further, the at least one backlight, in the HSL color space, may have a L coordinate of 0.13 < L < 0.78. The at least one backlight, in the HSL color space, may have a H coordinate of 338 < H < 342, a S coordinate of 0.3 < S < 1 .0 and a L coordinate of 0.65 < L < 0.70. For example, the at least one backlight, in the HSL color space, has a H coordinate of 340, a S coordinate of 0.75 and a L coordinate of 0.68. This specific red backlight may correspond to a RGB color of (R,G,B)=(234 / 108 / 150). Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 340, a S coordinate of 0.78 and a L coordinate of 0.68. This specific red color may correspond to a RGB color of (R,G,B)=(255,87,143). Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 340, a S coordinate of 0.71 and a L coordinate of 0.68. This specific red color may correspond to a RGB color of (R,G,B)=(231 ,111 ,151 ). Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 340, a S coordinate of 0.71 and a L coordinate of 0.68. This specific red color may correspond to a RGB color of (R,G,B)=(231 ,111 ,151). Alternatively or additionally, the at least one backlight, in the HSL color space, may have a H coordinate of 0, a S coordinate of 1 .0 and a L coordinate of 0.43. This specific red color may correspond to a RGB color of (R,G,B)=(219,0,0). These specific red backlights may achieve good purity and high product yield, as will be outlined in further detail below.

[0062] In one embodiment of this invention, for the low L or low L+S methods described herein (to shoot out darker colored, such as dark blue, seeds), the backlights in the HSL color space as described herein may be used, but also a white LED backlight or a backlight with an L coordinate of L = 255 (white). Hence, whenever a reference is made to a backlight in the low L / light- ness method or the low L+S method as described herein, such a white LED backlight or a backlight with an L coordinate of L = 255 (white) can be used as alternative backlight.

[0063] As outlined above, step ill. comprises automatically identifying, from the image taken in step ii. , seeds to be sorted out from the seed stream. In step ill., seeds to be sorted out from the seed stream may be colored seeds identified by identifying, in the image, objects cumulatively fulfilling the following conditions: the objects have predefined color coordinates, specifically color coordinates in a predefined range in the HSL color space, and the objects have one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape. Specifically, the identification of the seeds to be sorted out may comprise determining color coordinates and an area of the objects in the image, wherein the objects are determined to have the predefined color coordinates if the determined color coordinates are within a predefined range in the HSL color space; and the objects are determined to have the predefined area if the determined area exceeds an area threshold.

[0064] The predefined range, in the HSL color space, may have a H coordinate from 0 to 85, specifically from 5 to 65 and / or from 19 to 79, a S coordinate from 15 to 85, specifically from 18 to 82 and / or from 19 to 83, a L coordinate from 60 to 150, specifically from 91 to 141 and / or from 66 to 130, wherein the area threshold may be in the range of 200 to 1500 such as 400 to 1000 or 500 to 1000 pixels, specifically in the range of 600 to 850 pixels, more specifically in the range of 650 to 700, most specifically is 700 pixels (generally, for smaller seeds the pixel size is best reduced so as to ensure that a relevant area is covered). For example, objects in the image having a H coordinate in the range of 19 to 79, a S coordinate in the range of 19 to 83 and a L coordinate in the range of 66 to 130 may be identified as seeds to be sorted out from the seed stream in case the respective object may have a size of more than 700 pixels. Alternatively or additionally, objects in the image having a H coordinate in the range of 5 to 65, a S coordinate in the range of 18 to 82 and a L coordinate in the range of 91 to 141 may be identified as seeds to be sorted out from the seed stream in case the respective object may have a size of more than 700 pixels. The HSL coordinates may be given on an absolute scale from 0 to 255 for L or S values, which can be also transformed into a relative scale from 0 to 1 , as will be apparent to the skilled person, and on an absolute scale from 0 to 360 for H value.

[0065] Alternatively or additionally, the identification of the seeds to be sorted out may comprise determining at least one recognition parameter comprising weighting the color coordinates with the area of the objects in the image, specifically by using a product of the color coordinates and the area of the objects in the image. Thus, in case the recognition parameter exceeds a certain threshold, the object in the image may be identified as seeds to be sorted out from the seed stream.

[0066] In one embodiment, the method of the invention may comprise the determination of the object settings (as provided by the sorting device) for the darkest colored seed lot available, and the determination of the object settings (as provided by the sorting device) for the lightest colored seed lot available. Both of these object settings are being combined as 2 different rules into one sorting protocol and tested against the “white seeds” of the seedlots. The settings of L min of the light blue rule might be raised and / or the pixel size of both rules might be adapted to ensure no (or very limited) shoot out of white seeds and an effective shoot out of all blue seeds from any blue / white mixture of seeds. For example, this method can be performed on a batch of seed containing dark and light blue aleurone seeds and non-colored “white” seeds, with the same protocol of sorting out objects in the image having (1 ) a H coordinate in the range of 5 to 65, a S coordinate in the range of 18 to 82 and a L coordinate in the range of 88 to 145 and a size of more than 700 pixels (for sorting-out light blue seeds) and objects in the image having (2) a H coordinate in the range of 19-79, a S coordinate in the range of 19-83 and a L coordinate in the range of 62 to 130 and a size of more than 700 pixels (for dark blue seeds), preferably using an HSL backlight setting with a Hue value of 340, 71 or 0, such as any of the HSL backlight settings of : H340 / S192 / L172, H71 / S255 / L123, or H0 / S255 / L110. In one embodiment of this method, said image has a H coordinate in (1) and (2) comprising the range of 354<H<45 (e.g., the entire Hue range, or from 354<H to H<45. These two object specifications (1) and (2) can be combined to shoot out light blue (LB) seeds and dark blue (DB) seeds (this is called the LB_DB or DB_LB Shout Out method). The method can be performed once or can be repeated several times, such as repeated 1-3 times (hence, doing the sorting for 2-4 times in total), e.g., to maximize non-colored seed purity.

[0067] The method may specifically be a continuous method. The term “continuous” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a property of a process of being uninterrupted in time. Specifically, the continuous method of sorting seeds, such as cereal seeds, may comprise performing the methods steps repeatedly and at least partially overlapping in time. In step i., a continuous seed stream may be supplied to the sorting station. In step ii ., a continuous stream of images may be taken of the seed stream. In step ill., the stream of images may be continuously evaluated for continuously identifying seed to be sorted out from the seed stream.

[0068] In the method, a batch of seeds may be provided. The term “batch” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to an assembly of jointly produced objects or an assembly of objects produced at different times and / or places that were mixed. Specifically, the batch of seeds may comprise a plurality of seed which were jointly produced, specifically with respect to at least one of production place and production time. The batch of seeds may be subjected to method steps i.-iv. repeatedly, specifically at least twice. In each repetition, the batch may be diminished by the seeds ejected in step iv. of the previous run. Thus, by repeating performing method steps i.- iv. on the batch of seeds, the purity of seeds passing the sorting station may be enhanced.

[0069] The method of the current invention may further comprise at least one backlight identification step, specifically at least one backlight identification step preceding step i.. The backlight identification step may comprise identifying the at least one backlight to be used for sorting. Thus, in the backlight identification step, the at least one backlight used for sorting in steps i. to iv. may be determined. The backlight identification step may comprise a plurality of sortings with a plurality of backlights, specifically of different backlights. Each sorting may comprise using an aliquot of the seeds to be sorted. The at least one backlight may be determined via at least one quality control step. The quality control step may comprise evaluating a purity of the sorting across the plurality of backlights, and optionally further considering a product yield of the sorted seeds by weighing the corresponding aliquot fractions. Alternatively, the quality control step may comprise evaluating a purity of the sorting and an amount of losses of desired seeds of the sorting across the plurality of backlights. This method can be advantageously used to further improve sorting efficiency for a large-scale seed production using the specific backlight (within the HSL / RGB ranges as provided herein) that is best suited for that specific seed production.

[0070] The method may specifically comprise performing steps i. to iv. with at least one first backlight, the first backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The method may further comprise repeating steps i. to iv. using one of an ejected fraction of seeds or a retained fraction of seeds with either said first backlight or with at least one second backlight, the second backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The second backlight may be different from the first backlight. Thus, for example, the method may comprise at least two consecutive sorting rounds, wherein a first sorting round may comprise using the first backlight and the second sorting round may comprise using the second backlight, the second sorting round being performed using the ejected fraction of seeds or the retained fraction of seeds. Further, a third or even more sorting rounds may be performed with the other of the ejected fraction of seeds or the retained fraction of seeds, the third round of sorting may comprise using the same first backlight or another further backlight as will be outlined in further detail below.

[0071] The first backlight may be used for a first purpose of sorting seeds, such as cereal seeds. The first purpose of sorting (cereal) seeds may comprise at least one purpose selected from the group consisting of: maximize yield of non-colored (cereal) seeds; maximize yield of (darker) colored (cereal) seeds; maximize purity of non-colored or lighter colored (cereal) seeds, specifically by removing of impurities, such as (darker) colored seeds, weed seeds, disease-infected seeds, fungal sclerotia formed on the ears (such as ergot fruiting bodies) and / or other plant parts; maximize purity of (darker) colored (cereal) seeds; reduce number of (darker) colored seeds in non-colored or lighter colored seeds; reduce number of lighter colored or non-colored seeds in (darker) colored seeds; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof.

[0072] The second backlight may be used for a second purpose of sorting seeds, such as cereal seeds. The second purpose of sorting (cereal) seeds may comprise at least one purpose selected from the group consisting of: maximize yield of non-colored or lighter colored (cereal) seeds; maximize yield of (darker) colored (cereal) seeds; maximize purity of non-colored or lighter colored (cereal) seeds; maximize purity of (darker) colored (cereal) seeds; reduce number of (darker) colored seeds in non-colored or lighter colored seeds; reduce number of non-colored or lighter colored seeds in (darker) colored seeds; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof. The first backlight may be used for a first purpose of sorting cereal seeds with a blue aleurone or sorting seeds lacking a blue aleurone. The first purpose of sorting cereal seeds may comprise at least one purpose selected from the group consisting of: maximize yield of non-colored cereal seeds lacking a blue aleurone; maximize yield of colored cereal seeds with a blue aleurone; maximize purity of non-colored cereal seeds lacking a blue aleurone, specifically by removing of impurities, such as colored seeds with a blue aleurone, weed seeds, disease-infected seeds and / or other plant parts; maximize purity of colored cereal seeds with a blue aleurone; reduce number of colored seeds with a blue aleurone in non-colored seeds lacking a blue aleurone; reduce number of non-colored seeds lacking a blue aleurone in colored seeds with a blue aleurone; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of blue aleurone seeds while reducing losses of blue aleurone seed; or any combinations thereof.

[0073] The second backlight may be used for a second purpose of sorting cereal seeds with a blue aleurone or sorting seeds lacking a blue aleurone. The second purpose of sorting cereal seeds may comprise at least one purpose selected from the group consisting of: maximize yield of non-colored cereal seeds lacking a blue aleurone; maximize yield of colored cereal seeds with a blue aleurone; maximize purity of non-colored cereal seeds lacking a blue aleurone, specifically by removing of impurities, such as colored seeds with a blue aleurone, weed seeds, disease- infected seeds and / or other plant parts; maximize purity of colored cereal seeds with a blue aleurone; reduce number of colored seeds with a blue aleurone in non-colored seeds lacking a blue aleurone; reduce number of non-colored seeds lacking a blue aleurone in colored seeds with a blue aleurone; maximize purity of non-colored seeds while reducing losses of non-colored seed, maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof.

[0074] Specifically, the second purpose of sorting seeds, such as cereal seeds, may be different from a first purpose of sorting seeds, such as cereal seeds.

[0075] A repetition of steps i. to iv. using one of the ejected fraction of seeds or the retained fraction of seeds may comprise using a different sorting protocol compared to an initial sorting in step ill.. The sorting protocol may define a parameter according to which the seed to be sorted out is identified in the image. Thus, as an example, in step ill., seeds to be sorted out from the seed stream may be identified by identifying, in the image, objects having color coordinates of a predefined sorting protocol, specifically color coordinates in a predefined range in the HSL color space. This can be used to sort out dark and light colored seeds from non-colored seeds, such as sort out (dark and light) blue aleurone seeds from non-colored seeds, or to sort out dark colored seeds from light colored or non-colored seeds, such as sort out 3n dark blue aleurone seeds from light blue aleurone seed or non-colored seeds, sort out light colored seeds from dark colored and non-colored seeds, such as sort out light blue aleurone seeds from dark blue aleurone and non-colored seeds. In the following, HSL object settings (for seeds to be shot-out) are provided on an absolute scale but can equivalently also be transformed to a relative scale, as outlined above. The predefined sorting protocol may comprise at least one protocol with HSL object settings for the seeds to be ejected / shot-out selected from the group consisting of : a H coordinate comprising the range of 354<H<45 or the range of H > 354 and H < 54, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax<99; a H coordinate comprising the range of 354<H<45 or in the range of H > 345 and H < 79, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin <30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax^99; a H coordinate comprising the range of 354<H<45 or in the range of H > 345 and H < 45, a S coordinate in the range of Smin < S < 98, wherein Smin is in the range of 55 < Smin 62 and a L coordinate in the range of Lmin < L < 140, wherein Lmin is in the range of 96 < Lmin 102; a H coordinate comprising the range of 354<H< 45 or in the range of H > 354 and H < 54, a S coordinate in the range of Smin < S < 98, wherein Smin is in the range of 55 < Smin 62 and a L coordinate in the range of Lmin < L < 140, wherein Lmin is in the range of 96 < Lmin 102; a combined protocol with first color coordinates comprising a H coordinate comprising the range of 354<H<45 or in the range of 19 < H < 79, a S coordinate in the range of 19 < S < 83, and a L coordinate in the range of 62 < L < 130, and with second color coordinates comprising a H coordinate comprising the range of 354<H<45 or in the range of 5 < H < 65, a S coordinate in the range of 18 < S < 82, and a L coordinate in the range of 88 < L < 145; a H coordinate comprising the range of 354<H<45, a S coordinate in the range of 35 < S < 99, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0< Lmin— 30 and wherein the Lmax is set so as to get a certain % shot-out, such as an Lmax of 78<Lmax^127; a H coordinate comprising the range of 354<H<45, an S coordinate in the range of Smin— S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L< Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax^99; a H coordinate comprising the range of 354<H<45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 96<Lmin^110, 96<Lmin^102 or 97<Lmin^110, such as an Lmin of 105, and the Lmax is

[0076] 138<Lmax^ 145, such as an Lmax of 140, and an S coordinate in the range of Smin^S<Smax wherein the Smin is 55<Smin<75, 55<Smin<62 or 65<Smin<75, such as an Smin of 70, and the Smax is 95<Smax^255, such as an Smax of 98 or 120; or said predefined sorting protocol may comprise at least one protocol with HSL object settings for the seeds to be ejected / shot-out, wherein the H object setting can be any H range, as long as it includes 354 < H < 45, such as a sorting protocol with H object settings for the seeds to be ejected / shot-out of any H range, but including the range of 354 < H < 45, the Smin and Smax as provided by the sorting device for the seeds to be shot-out, and an Lmin of 0 or 0<Lmin^30 and an Lmaxbetween 85 and 127.

[0077] Additionally, the seeds to be sorted out may further be identified by identifying objects in the image having one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape. For example, the seeds to be sorted out may be further identified by identifying objects in the image having a determined area exceeding an area threshold in the range of 200 to 1500 pixels or 300 to 1500 pixels, such as 400-1000 pixels, or in the range of 500 to 1000 pixels, specifically in the range of 550 to 800 pixels, more specifically in the range of 700 to 800, most specifically the area threshold is 200, 400, 600, 700, or 750 pixels. The pixel range for larger seeds with an average thousand kernel weight (TKW) above 40 may be generally preferably higher (700-1500 pixel) as compared to seed lots with an average TKW around 30 (300-600 pixel), while smaller pixel sizes (such as 200) is for shooting out seeds with black spots. Also, sieving seeds or separating seeds based on size / weight, such as before the sorting method of the invention, can help improve sorting accuracy and reduce losses.

[0078] The method may specifically comprise separately repeating steps i. to iv. with the ejected fraction of seeds and the retained fraction of seeds. For example, the steps i. to iv. may be repeated (once or more, such as 1-3 times) using the retained fraction of seeds with the first backlight, wherein the steps i. to iv. may be repeated using the ejected fraction of seeds with the second backlight. The retained fraction may mostly contain the non-colored seeds, such as sterile “white” seeds lacking a blue aleurone, and the ejected fraction may mostly contain the colored seeds, such as fertile seeds having a blue aleurone. In the following, the first backlight and the second backlight and further optional backlights may be given in absolute HSL values (in a 360 scale for the H and a 255 scale for the S and L values). However, these color values may be equivalently transformed to relative HSL values or even to different color spaces, e.g. to the RGB color space, as outlined above. In the above, the first backlight, in the HSL color space, may have at least one color selected from the group consisting of: a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123. The second backlight, in the HSL color space, may have at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110, or any H coordinate differing from the above H coordinates by + / - 3 (in the 0-360 scale), any S coordinate differing from the above S coordinates by + / - 15 (in the 0-255 scale), or any L coordinate differing from the above L coordinate by + / - 5 (in the 0-255 scale).

[0079] Alternatively or additionally, the steps i. to iv. may be done using a first backlight that may have, in the HSL color space, a color with, e.g., a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110 (e.g., to maximize the yield of non-colored (such as sterile “white”) seeds, and the steps i. to iv. may be repeated using the retained fraction of seeds with a second backlight that differs from said first backlight (and that may have, in the HSL color space, e.g., at least one color having a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172 or a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123), wherein the steps i. to iv. may be repeated using the ejected fraction of seeds with a third backlight, the third backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The third backlight may be different from the first and second backlight. In this example, the first backlight, in the HSL color space, may have a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110, or has a H coordinate of 71 , a S coordinate of 255 and a L coor- dinate of 123. The second backlight, in the HSL color space, may have at least one color selected from the group consisting of: a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123. Any of the second backlights may be used to repeat at least once (such as 1-3 times) the steps i. to iv. using the retained fraction of seeds. The third backlight, in the HSL color space, may have at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123.

[0080] Alternatively or additionally, the steps i. to iv. may be repeated using the retained fraction of seeds with a second backlight (that may have a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172), wherein the steps i. to iv. may be repeated using the ejected fraction of seeds with the first backlight (that may have a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110). Alternatively or additionally, in the above method using the first backlight (that may have a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110), the steps i. to iv. may be repeated using the ejected fraction of seeds (e.g., mostly dark and light blue seeds) with the first backlight (that may have a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110), wherein the steps i. to iv. may be repeated using the ejected fraction of seeds from the repetition using the ejected fraction of seeds (e.g., mostly dark and light blue seeds) with a further third backlight (that may have a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; or a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110; or a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123).

[0081] The method may further comprise repeating steps i. to iv. with a retained or an ejected fraction of seeds from a repetition of steps, i. to iv. using the ejected fraction of seeds, wherein the further repetition uses a fourth backlight, the fourth backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The fourth backlight may be different from the first and / or second backlight. The fourth backlight, in the HSL color space, may have at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123.

[0082] A fraction of ejected and / or retained seeds comprising, specifically consisting of, non-colored seeds, such as cereal seeds, may be added to a repetition of steps i. to iv. using the retained fraction of seeds. Indeed, e.g., retained seeds obtained after shooting out colored (such as blue aleurone) seeds in the method aiming for purer non-colored (such as male-sterile seeds, in the non-colored seed enrichment stream) can be used in a further shooting out of colored (such as blue aleurone) seeds to maximize non-colored seed purity, and also the retained seeds obtained after shooting out light and dark colored (such as light and dark blue aleurone) seeds in the colored seed stream (such as the 1 n / 2n blue aleurone maintainer stream) can be used in or added to the seeds used in the non-colored (such as male sterile) seed sorting stream to maximize yield of non-colored (such as male-sterile) seeds, certainly when handling high volumes of seed in large-scale productions in the field.

[0083] The repetition of steps i. to iv. using the retained fraction of seeds may be repeated at least twice.

[0084] Whenever reference is made herein to a specific backlight setting in the HSL color space by reference to a specific numeric value for the H, S or L backlight setting in this application, such as the exemplified specific HSL settings or the specific HSL settings referred to in the description and / or the Figures, then said specific HSL setting can differ to the specific value given by + / 3, + / - 2 or + / - 1 for the H value (on the 0-360 scale), by + / - 15, + / - 10, or + / - 5 for the S value (on the 0-255 scale) or by + / - 5, + / - 4, + / 3, + / - 2 or + / - 1 for the L value (on the 0-255 scale). Hence, an H value of 20 can be replaced by an H value of 17 or 23 or any value between 17 and 23, an S value of 192 can be replaced by an S value of 177 or 207 or any value in between 177 and 207, an L value of 123 can be replaced by an L value of 118 or 128 or any value in between 118 and 128. Also, before using specific HSL backlight settings on another seed sorter device, the values may need to be adjusted depending on the quality of the backlight, camera, etc. used, to get a similar sorting efficiency. Also, in some cases an alternative HLS backlight can be used in a certain step (such as those illustrated in the description of Figures 4-9).

[0085] In the above, in the sorting steps of the seed stream improving purity of non-colored (such as male sterile seeds lacking a blue aleurone) seeds (non-colored seed stream), the first backlight in the first sorting step can be used to retain non-colored seeds and shoot-out colored seeds (such as using the DB_LB shoot out method described herein with HSL backlights H340 / S192 / L172 or H71 / S255 / L123 to maximize the yield of colored (such as blue aleurone) seeds ejected, as can be seen in Figs. 4-7 (left part)), or using the DB_LB shoot out method described herein with the HSL backlight H0 / S255 / L110 to maximize the yield of non-colored seeds (such as male-sterile seeds without blue aleurone) in the retained fraction (as can be seen in Figs. 8 and 9 (left part)). In a second sorting step, in the non-colored seed stream (such as the male sterile seed enrichment seed stream) using the seeds retained in the first sorting step using the first backlight, the first backlight can be used repeatedly (such as repeated 1 time, 2 times or 3 times, or 1-3 times) to purify non-colored seeds (such as the male-sterile seeds) and shoot-out colored seeds (such as dark and light blue seeds), when the first backlight has HSL settings H340 / S192 / L172 or H71 / S255 / L123 (such as by using the DB_LB Shoot out method described herein with the (same or the alternative) first backlight as in the first sorting step, as can be seen in Figs. 4-7 (left part)). Alternatively, in a second sorting step in the non-colored seed stream using the seeds retained with the first sorting step using the first backlight, a second backlight can be used to purify non-colored seeds (such as the male-sterile seeds) and shoot-out colored seeds (such as dark and light blue seeds), when the first backlight has HSL settings H0 / S255 / L111 , such as wherein said second sorting step uses the DB_LB Shoot out method described herein with the second backlight HSL settings H340 / S192 / L172 or H71 / S255 / L123, as can be seen in Figs. 8 and 9 (left part). In the above, the ejected / shot-out colored seeds from the first sorting step using said first backlight can be used in the seed stream improving the purity of colored (such as blue or light blue (1 n / 2n maintainer)) seeds (the colored seed stream). Hence, said first sorting step in the noncolored seed stream is also the first sorting step in the colored seed stream, using the same first backlight as in the non-colored seed stream (but instead of the retained seeds, the ejected seeds are used in the colored seed stream). In a second sorting step in said colored seed stream, the ejected colored (such as blue, fertile (maintainer) seed) seeds of the first sorting step using the first backlight, are depleted from dark colored (such as 3n dark blue) seeds using a second backlight to purify lighter colored (such as 1 n / 2n blue seeds) by shooting-out the dark colored seeds (such as 3n dark blue seeds), wherein the second backlight has HSL settings H20 / S255 / L150, H111 / S255 / L123 or H0 / S255 / L110 (such as by using the dark blue (DB) Shoot out method described herein with that second backlight (which can use the low L or low L+S method as described herein), as can be seen in Figs. 4-8 (right part)), or the second backlight in said second sorting step has HSL settings H0 / S255 / L110 (such as by using the DB_LB Shoot out method described herein with that second backlight (as can be seen in Fig.9 (right part)).

[0086] Also, in said colored seed stream a third sorting step can be done using a third backlight to either : i) shoot out light colored (such as light blue) seeds (such as by using the LB Shoot out method described herein with a third backlight with HSL settings H0 / S255 / L111 , H71 / S255 / L123 or H20 / S255 / L150, as can be seen in Figs. 4 and 5 (right part), or with a third backlight with HSL settings H0 / S255 / L111 or H71 / S255 / L123, as can be seen in Fig. 8 (right part)), or ii) shoot out light and dark colored seeds (such as blue and dark blue seeds, such as by using the LB_DB Shoot out method described herein with the third backlight with HSL settings H340 / S192 / L172 or H71 / S255 / L123, as can be seen in Fig. 6 (right part), or by using the LB_DB Shoot out method described herein with the third backlight with HSL settings H0 / S255 / L110 or H71 / S255 / L123 (as can be seen in Fig. 7(right part)), or iii) shoot out dark colored (such as 3n dark blue ) seeds, such as by using the DB Shoot out method described herein (which can use the low L or low L+S method as described herein) with the third backlight with HSL settings H20 / S255 / L150, H0 / S255 / L110 or H111 / S255 / L123 (as can be seen in Fig. 9 (right part)). In the above, included are also any backlight with an H coordinate differing from the above H coordinates by + / - 3 (in the 0-360 scale), any S coordinate differing from the above S coordinates by + / - 15 (in the 0-255 scale), or any L coordinate differing from the above L coordinate by + / - 5 (in the 0-255 scale).

[0087] In one embodiment of the above methods, besides the preferred backlight HSL settings as described above, also the L object settings, or the L and S object settings for the objects to be ejected are modified from those provided by the sorting device, such as by lowering the Lmin and adapting the Lmax to a certain quantity of seeds to be shot-out, or lowering the Lmin and Smin, and adapting the Lmax or the Lmax and Smax to a certain quantity of seeds to be shot-out (e.g., the H, S, and L object settings as specifically described herein for the low L (lightness) or low L+S methods to remove dark blue seeds, or as specifically described for the light blue (LB) Shoot out method to retain dark colored and non-colored seeds). Also, in the above schemes, such as in the schemes in Figures 4 to 9, besides the sorting according to the method as described herein, using adapted backlight HSL settings (with or without reduced / adapted L or L and S object settings), also any other sorting method or tool can be used that can help improve to maximize yield of non-colored (cereal) seeds; maximize yield of colored (cereal) seeds; maximize purity of non-colored (cereal) seeds, specifically by removing of impurities, such as colored seeds, weed seeds, disease-infected (such as ergot-infected) seeds and / or other plant parts; maximize purity of colored (cereal) seeds; reduce number of colored seeds in non-colored seeds; reduce number of non-colored seeds in colored seeds; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof. In this regard, in addition to the methods described herein, dark blue 3n aleurone seed, light blue 1 n / 2n blue aleurone seed, or non-colored “white” seed removal or retention can be done by visual and / or infrared spectral analysis, specifically near-infrared spectral analysis, and / or UV spectral analysis, and / or spectral analysis using x-rays, and / or spectral analysis using Raman scattering, with or without using a trained neural network to improve sorting efficiency. Any other (color) sorting method or other sorter (e.g., the Cimbria SEA.IQ PLUS, or a SMART sorter from Any- Sort), such as a method or sorter using artificial intelligence (provided with or without adaptable backlight color) can be used next to (before, during or after) the methods of the invention to shoot-out any (remaining) non-colored “white” seeds from the blue / colored seed stream. In one embodiment, such other sorting method, may be done on / by the same sorting device as the method(s) of the invention.

[0088] The method may be for sorting out colored seeds from a seed mixture containing darker and lighter colored seeds and non-colored seeds, wherein a sorting protocol may use a combination of a) the HSL object settings for the darker colored seeds as provided by the sorting device and b) the HSL object settings for the lighter colored seed as provided by the sorting device, which method can be performed once or can be repeated several times, such as repeating 1-3 times, to maximize non-colored seed purity.

[0089] In the method, L object settings as provided by the sorting device for the darker colored seeds, may be modified by lowering the Lmin object settings to at least the value of the darkest seed to ensure a shoot-out of the darkest seeds, and the Lmax may be set to a value resulting in the shoot-out of a desired % by weight of the darkest seeds, and wherein the H object setting is any H range comprising 354<H<45, such as wherein the Lmin is 0 or 0<Lmin^60 or 0<Lmin^30.

[0090] At least step ill., and optionally at least one of steps II. and iv., may specifically be at least one of computer-implemented and computer-controlled. The term “computer-implemented” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a method or method step involving at least one computer and / or at least one computer network. The computer and / or computer network may comprise at least one controller which is configured for performing at least one of the method steps of the method according to the present invention. The computer-implemented method steps may be performed completely automatically, specifically without user interaction. The term “computer-controlled” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a method or method step involving control by at least one computer and / or at least one computer network. Specifically, the computer-controlled method steps may involve using at least one computer and / or computer network which may comprise at least one controller being configured for controlling at least one of device, such as the sorting station and / or any parts thereof, to perform the specific task or function. For example, at least one computer and / or computer network may comprise at least one controller being configured for controlling the camera to take the image of the backlighted seed of the seed stream. Alternatively or additionally, at least one computer and / or computer network may comprise at least one controller being configured for controlling the sorting station, specifically the ejector, to eject seeds identified to be sorted out from the seed stream.

[0091] In step ill, from the image taken in step II, the automatic identification from the seeds to be sorted out from the seed stream may be a trained artificial neural network (ANN). The term “artificial neural network”, also referred to as “neural network”, as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a mathematical model comprising of a plurality of units or nodes which are connected by one or more edges. Each of the units or nodes may receive at least one input signal from one or more connected units or nodes and may be configured for processing the at least one input signal and forward at least one output signal to one or more connected units or nodes. The input signal and / or the output signal may be a real number. The output signal of each unit or node may be computed by at least one mathematical function, specifically at least one non-linear function, taking into account the sum the input signals. The mathematical function of the unit or node may be an activation function of the unit or node. A strength of the input signal at each connection may be determined by using an adjustable weight, wherein the adjustable weight may be adjusted during a learning or training process. The units or nodes may be aggregated into two or more layers. An input signal may be given to an input layer and may be forwarded to an output layer. Optionally, the artificial neural network may comprise one or more hidden layers in between the input layer and the output layer.

[0092] The method may further comprise at least one training step. In the training step, the ANN may be trained using labeled images. The labeled images may comprise genotyping data and / or provide information on seeds having a normal (non-colored) aleurone and cereal seeds containing a blue aleurone. Thus, the labeled images may provide ground truth data for the training step.

[0093] The trained ANN may be trained using records of training data. A record of training data may comprise training input data and corresponding training output data. The training output data of a record of training data may be the result that is expected to be produced by the ANN when being given the training input data of the same record of training data as input. The deviation between this expected result and the actual result produced by the ANN may be observed and rated by means of a “loss function”. This loss function may be used as a feedback for adjusting the parameters of the ANN. For example, the parameters may be adjusted with the optimization goal of minimizing the values of the loss function that result when all training input data is fed into the ANN and the outcome is compared with the corresponding training output data. The result of this training may be that given a relatively small number of records of training data as “ground truth”, the ANN is enabled to perform its job well for a number of records of input data higher by many orders of magnitude. Thus, the ANN may comprise at least one algorithm and model parameters. Parameters of the ANN may be adjusted in the training step.

[0094] The method may further comprise at least one second spectral seed sorting step. The term “spectral seed sorting step” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a process of sorting seed by spectral means. Specifically, the second spectral seed sorting step may comprise using at least one spectrometer device, such as a device configured for acquiring at least one optical property or optically measurable property as a function of a wavelength. In particular, the spectrometer device used in the second spectral seed sorting step may be or may comprise an apparatus configured for recording a signal intensity with respect to the corresponding wavelength of a spectrum or a partition thereof, such as a wavelength interval. The second spectral seed sorting step may comprise one or more of an infrared spectral analysis, specifically a near-infra- red spectral analysis, a UV spectral analysis, a spectral analysis using x-rays and / or a spectral analysis using Raman scattering. The second spectral seed sorting step, such as a near-infra- red spectral analysis, may be performed simultaneously or consecutively with steps i. to iv.. For example, the second spectral seed sorting step may be performed after steps i. to iv., e.g. as a subsequent sorting round to steps i. to iv.. Alternatively, the spectral seed sorting step may be performed in a timely overlapping fashion with steps i. to ill., wherein the ejecting in step iv. may be performed on the basis of the outcome of step ill. and / or the second spectral sorting step.

[0095] The second spectral seed sorting step may comprise determining at least one item of spectroscopic information on the seeds of the seed stream, that is different from the information determined in the above first visual light sorting step, specifically in method steps i. to. ill.. The item of spectroscopic information may comprise at least one of a transmission, an absorption, a reflection and an emission of the seeds in the seed stream with respect to the corresponding wavelength of a spectrum or a partition thereof, such as a wavelength interval. The second spectral seed sorting step may comprise determining at least one a transmission, an absorption, a reflection and an emission in the infrared spectral range, e.g. in a wavelength range from 760 nm to 1000 pm, specifically in a near-infrared spectral range, e.g. in a wavelength range from 760 nm to 1 .5 pm. The at least one item of spectroscopic information may be used for automatically identifying seeds to be sorted out from the seed stream. For example, the item of spectroscopic information may be used for determining one or more of a presence, an absence and a concentration of a specific compound in the seeds, e.g. of an organic colorant, such as anthocyanin. Methods for identifying anthocyanin contents by NIR spectroscopy are known to the skilled person, e.g. as described in Stuppner et al. (2020, Sensors 20(17), 4983 (https: / / doi.org / 10.3390 / s20174983), Chen et al. (2015, Food Chemistry 172, 788-793 (https: / / doi.Org / 10.1016 / j.foodchem.2014.09.119), or WO 2023 / 088892 A2. The seeds in the seeds stream for which one or more of a presence, an absence and a concentration of the specific compound is detected may be ejected by using the ejector of the sorting station.

[0096] The second spectral seed sorting step may further comprise a trained artificial neural network which automatically identifies seeds to be sorted out from the seed stream based on the at least one item of spectroscopic information. The method may further comprise at least one training step. In the training step, the ANN may be trained using labeled spectroscopic information. The labeled spectroscopic information may comprise genotyping data and / or provide information on seeds having a normal (non-colored) aleurone and cereal seeds containing a blue aleurone. Thus, the labeled spectroscopic information may provide ground truth data for the training step.

[0097] The method may further comprise at least one control analysis step. The control analysis step may comprise using a multispectral imaging device having black-colored background, such as a black-colored conveyor belt. Specifically, the multispectral imaging device may have a blackcolored conveyor belt. The black-colored conveyor belt may specifically give the most accurate control results with respect to visual inspection.

[0098] In a further aspect of the present invention, a sorting device for sorting seeds is disclosed. The term “sorting device” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a device configured for performing at least one sorting function. Specifically, the sorting device may be configured for differentiating and separating the seeds to be sorted out from the other seeds in the seed stream. The sorting device comprises, as will be outlined in further detail below, at least one seed feeder for supplying a seed stream. The sorting device may be or may comprise an optical sorting device. Specifically, the sorting device may comprise at least one optical system which is configured, in conjunction with image processing software, for identifying seeds to be sorted out in the seed stream. For example, the camera and the backlight device may be part of the optical system of the sorting device. The sorting device may further comprise at least one separation system for performing separation of the seeds to be sorted out from the other seeds in the seed stream. For example, the ejector may be part of the separation system.

[0099] The sorting device comprises:

[0100] I. at least one seed feeder for supplying a seed stream to at least one sorting station; and

[0101] II. at least one sorting station, comprising at least one backlight device for backlighting seeds of the seed stream, the sorting station further comprising at least one camera for taking at least one image of the backlighted seed, and the sorting station further comprising at least one ejector for ejecting seeds from the seed stream. The sorting device is configured for performing the method according to the present invention, such as according to any one of the embodiments disclosed above and / or according to any one of the embodiments disclosed in further detail below. Thus, for definitions of terms and / or description of possible embodiments, reference is made to the description of the method of sorting seeds above.

[0102] The term “seed feeder” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a device configured for supplying a seed stream. Specifically, the seed feeder may comprise at least one feed hopper configured for receiving a plurality of seeds and for providing the seeds to at least one chute in a controllable fashion. The seed feeder may further comprise at least one vibratory feeder configured for applying vibrations to the feed hopper such that seeds comprised therein may leave the feed hopper to the at least one chute. The seed feeder may be configured for supplying the plurality of seeds arranged in a regular fashion. The seed feeder may be configured for supplying the plurality of seeds arranged individually in a row or line.

[0103] The sorting device may further comprise at least one controller. The term “controller” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to an arbitrary logic circuitry configured for performing basic operations of a computer or system, and / or, generally, to a device which is configured for performing calculations or logic operations. In particular, the controller may be configured for processing basic instructions that drive the computer or system. As an example, the controller may comprise at least one arithmetic logic unit (ALU), at least one floating-point unit (FPU), such as a math co-processor or a numeric co-processor, a plurality of registers, specifically registers configured for supplying operands to the ALU and storing results of operations, and a memory, such as an L1 and L2 cache memory. In particular, the controller may be a multi-core processor. Specifically, the controller may be or may comprise a central processing unit (CPU). Additionally or alternatively, the controller may be or may comprise a microprocessor, thus specifically the controller’s elements may be contained in one single integrated circuitry (IC) chip. Additionally or alternatively, the controller may be or may comprise one or more application-specific integrated circuits (ASICs) and / or one or more field-programmable gate arrays (FPGAs) and / or one or more tensor processing unit (TPU) and / or one or more chip, such as a dedicated machine learning optimized chip, or the like. The controller specifically may be configured, such as by software programming, for performing one or more operations. The controller may be configured for performing at least step ill. of the method. Optionally, the controller may further be configured for controlling at least one of steps II. and iv. of the method.

[0104] The sorting device may further comprise at least one target chute and at least one sort-out chute. The term “target chute” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a seed passage dedicated to seeds which are not be sorted out. The term “sort-out chute” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to a seed passage dedicated to the seeds to be sorted out. The target chute and the sort-out chute may specifically be different from each other. The seeds ejected by the ejector may be collected by the sort-out chute. The remaining seed stream may be collected by the target chute. The sorting station, specifically the ejector, may be configured for separating the seeds from the seed stream into the sort-out chute in case the seeds are identified as seeds to be sorted out. The sorting device may be configured such that seeds from the seed stream which are not identified as seeds to be sorted out may pass the sorting station towards the target chute.

[0105] The sorting station may be configured such that the seed stream passes the sorting station between the backlight device and the camera.

[0106] The sorting station may comprise at least two cameras. The cameras may be configured for taking images of the backlighted seed from different angles in space. The sorting station may further comprise at least two backlight devices. Each backlight device may be assigned to a camera of the at least two cameras. The cameras may be configured for taking images of the backlighted seed from opposing directions. For example, the sorting station may comprise at least one first backlight device assigned to at least one first camera. The first backlight device may be arranged on a first side with respect to the seed stream. The first camera may be arranged on a second side opposing the first side with respect to the seed stream. Further, the sorting station may comprise at least one second backlight device and at least one second camera. The second backlight device may be arranged on the second side and the second camera may be arranged on the first side.

[0107] The term “ejector” as used herein is a broad term and is to be given its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term specifically may refer, without limitation, to an arbitrary device configured for ejecting seeds. The ejector may specifically be or may comprise at least one of a mechanical device and a pneumatic device. The ejector may comprise at least one of a mechanical ejector having at least one mechanical actor and a pneumatic ejector having at least one nozzle for directing an air jet at the seed to be sorted out from the seed stream. For example, the ejector may comprise the at least one pneumatic ejector configured for ejecting the seed to be sorted out from the seed stream by using compressed air directed via one or more nozzles to separate the seed to be sorted out from the seed stream.

[0108] In a further aspect of the present invention, a computer program is disclosed, comprising instructions which, when the program is executed by the sorting device according to the present invention, such as according to any one of the embodiments described above and / or according to any one of the embodiments disclosed in further detail below, specifically by the controller of the sorting device, causes the sorting device to perform at least step ill. of the method according to the present invention, such as according to any one of the embodiments described above and / or according to any one of the embodiments disclosed in further detail below, and, optionally, at least one of steps II. and iv. of the method.

[0109] Thus, specifically, one, more than one or even all of method steps i. to iv. as indicated above may be performed and / or controlled by using a computer or a computer network, preferably by using a computer program.

[0110] In a further aspect of the present invention, a computer-readable storage medium, specifically a non-transient computer readable medium, is disclosed, comprising instructions which, when the instructions are executed by the sorting device according to the present invention, such as according to any one of the embodiments described above and / or according to any one of the embodiments disclosed in further detail below, specifically by the controller of the sorting device, cause the sorting device to perform at least step ill. of the method according to the present invention, such as according to any one of the embodiments described above and / or according to any one of the embodiments disclosed in further detail below, and, optionally, at least one of steps II. and iv. of the method.

[0111] As used herein, the term “computer-readable data medium” specifically may refer to non-transi- tory data storage means, such as a hardware storage medium having stored thereon computerexecutable instructions. The computer-readable storage medium specifically may be or may comprise a storage medium such as a random-access memory (RAM) and / or a read-only memory (ROM). The computer-readable storage medium may be or may comprise at least one computer-readable data carrier.

[0112] Further disclosed and proposed herein is a computer program product having program code means, in order to perform and / or control the method according to the present invention in one or more of the embodiments disclosed herein when the program is executed on a computer or computer network. Specifically, the program code means may be stored on a computer-readable data carrier and / or on a computer-readable storage medium.

[0113] Further disclosed and proposed herein is a data carrier having a data structure stored thereon, which, after loading into a computer or computer network, such as into a working memory or main memory of the computer or computer network, may execute and / or control executing of the method according to one or more of the embodiments disclosed herein.

[0114] Further disclosed and proposed herein is a computer program product with program code means stored on a machine-readable carrier, in order to perform and / or control performing of the method according to one or more of the embodiments disclosed herein, when the program is executed on a computer or computer network. As used herein, a computer program product refers to the program as a tradable product. The product may generally exist in an arbitrary format, such as in a paper format, or on a computer-readable data carrier and / or on a computer- readable storage medium. Specifically, the computer program product may be distributed over a data network.

[0115] Finally, disclosed and proposed herein is a modulated data signal which contains instructions readable by a computer system or computer network, for performing and / or controlling the method according to one or more of the embodiments disclosed herein.

[0116] Referring to the computer-implemented aspects of the invention, one or more of the method steps or even all of the method steps of the method according to one or more of the embodiments disclosed herein may be performed and / or controlled by using a computer or computer network. Thus, generally, any of the method steps including provision and / or manipulation of data may be performed by using a computer or computer network. Generally, these method steps may include any of the method steps, typically except for method steps requiring manual work, such as providing the samples and / or certain aspects of performing the actual measurements.

[0117] In a further aspect of the present invention, a use of the sorting device according to the present invention, such as according to any one of the embodiments described above and / or according to any one of the embodiments disclosed in further detail below, is disclosed, for a purpose of use, selected from the group consisting of: sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds; sorting out of dark colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds (with or without non-col- ored seeds in said mixture); sorting out of lighter colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds, which mixture may contain non-colored seeds; sorting out of dark colored seeds and lighter colored seeds from a mixture of seeds containing dark colored seeds, lighter colored seeds and non-colored seeds; sorting out of colored cereal seeds from a mixture of cereal seeds containing non-colored cereal seeds and colored cereal seeds; sorting out disease-infected seeds with darker seed color from non-infected seed; sorting out of colored cereal seeds containing a blue aleurone from a mixture of cereal seeds containing non-colored cereal seeds and cereal seeds containing a blue aleurone; sorting out of 3n dark blue aleurone seeds from a mixture of seeds containing 1 n, 2n and 3n blue aleurone seeds (with or without non-colored seeds); sorting out of 1 n and 2n lighter blue aleurone seeds from a mixture of seeds containing 1 n, 2n and 3n blue aleurone seeds (with or without non-colored seeds in said mixture); sorting out of 1 n, 2n and 3n lighter and darker blue aleurone seeds from a mixture of seeds containing 1 n, 2n and 3n blue aleurone seeds and non-colored seeds; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 70 < H < 150; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 25 < H < 70; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 0 < H < 25 or H > 310. Specifically, using the sorting device according to the present invention for sorting out of colored cereal seeds containing a blue aleurone from a mixture of cereal seeds containing non-colored cereal seeds (not having a blue aleurone) and cereal seeds containing a blue aleurone using at least one backlight having, in the HSL color space, a H coordinate of 70 < H < 150, may enable sorting of the non-colored cereal seeds with high purity. Specifically, in the sorting, the white seeds may be retained and the blue seeds may be shot out and used, e.g. when maintainer seeds are desired and / or when looking for anthocyanin colorant in the seeds at higher amount. Also, the 3n bla seeds may be removed from the 1n and 2n blue seeds. Alternatively or additionally, any blue seeds may be removed from white seeds by shooting out light and dark blue seeds.

[0118] Using the sorting device according to the present invention for sorting out of colored cereal seeds containing a blue aleurone from a mixture of cereal seeds containing non-colored cereal seeds and cereal seeds containing a blue aleurone using at least one backlight having, in the HSL color space, a H coordinate of 25 < H < 70, may enable sorting with high efficiency, i.e. sorting the non-colored cereal seeds with good purity and high product yield.

[0119] Using the sorting device according to the present invention for sorting out of colored cereal seeds containing a blue aleurone from a mixture of cereal seeds containing non-colored cereal seeds and cereal seeds containing a blue aleurone using at least one backlight having, in the HSL color space, a H coordinate of 0 < H < 25 or H > 310, may enable a very good purity and good product yield in the non-colored cereal seeds for seed streams having cereal seeds containing a blue aleurone with dark blue phenotype of the colored seeds, specifically in the seeds having the 3n blue aleurone.

[0120] The method of sorting seeds and the sorting device according to the present invention may provide a large number of advantages over known methods and devices. Specifically, the method of sorting seeds, by using colored backlight, may improve sorting purity and sorting yield in sorting of seeds containing non-colored seeds and colored seeds.

[0121] The method may comprise an improved seed sorting based on color differences in mature plant seeds, specifically for wheat seeds. Different colors of mature plant seeds may be used to sort some type of seeds from others, such as seeds comprising a color gene as a screenable marker to sort for another gene closely linked to the color gene. The method may specifically be useful when another color of seed means a higher anthocyanin content. Additionally or alternatively, the method may specifically be used in the fields of hybrid breeding where the male and female plants have seeds of different color and the selfed male seeds are to be removed from the hybrid seeds using a difference in color in the seeds (e.g., when selfed male seeds and hybrid seeds are harvested together, such as in mixed planting of a proportion of male plants in between a majority of female plants to produce hybrid seed). For example, in some instances, no blue seed may be allowed in certain cereal seeds sold, while blue seeds with higher anthocyanin content may appear in cereals. The method may also reliably sort seeds even when the seed color intensity depends on the number of copies of the color locus, such as a BLA locus, and when it is challenging to sort the different color intensities from non-colored seeds. The method may even be used for separating seeds containing 1 , 2 or 3 copies of the color locus, such as the BLA locus, in their aleurone layer.

[0122] The method may specifically provide accurate color sorting of cereal grains containing a BLA locus by using colored backlight. In particular, the ability to sort the non-BLA fraction to a high purity of 99 % or higher, or of 99,7% or higher may be dependent on the use of particular background light settings. In the method with the color values, in the HSL color space, of a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, such as for either yellow, red and / or green backlight, the purity in sorting may be above the threshold of 99 %, specifically of above 99.7 %, more specifically obtained by repeated shoot-outs, in particular when repeating the sorting for two or three times. Further, using differently colored backlight may influence the amount of losses occurring during the sorting into differently colored seed fractions. Also, in the method with the color values with the backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, such as for either yellow, red and / or green backlight, the losses of non-colored (“white”) seed (“white” seed ending up in the shoot-out fraction), when achieving a purity above 99 or 99.7 %, may be 4-45%, 4-40%, 4-35 %, 4-30 %, 4-25 %, 4-20 %, 4-15 %, or 4-10%. Using the method of the invention with multiple shoot-outs (without recovering white seeds from the maintainer (blue) seed stream), the sorting of the mixture of (fertile) blue aleurone and (male-sterile) non-colored seeds (produced in a good growing environment in a field trial in Germany) by 11 different blue aleurone wheat varieties gave on average 11 ,5 % (by weight) losses of white seed across these 11 varieties, varying from 4 to 33 % loss of “white” seed, and across those 11 different varieties, having an average purity of “white” seeds of 99,6 % (ranging from 99.2-99.8 %). Hence, using the method(s) of the invention with multiple shoot-outs, a loss of “white” seeds less than 45 %, less than 40 %, less than 35 %, less than 30 %, less than 25 %, less than 20 %, less than 15 %, or less than 10 % may be obtained, when a purity of at least 99 %, or at least 99,7 % of white seeds is obtained. Further, using the method of the invention with multiple shoot-outs (without recovering blue seeds from the male-sterile (“white”) seed stream), when sorting the mixture of (fertile) blue aleurone and (male-sterile) non-colored seeds as produced in a good growing environment in Germany in one field trial with 4 different varieties, the average loss of blue aleurone seeds (by weight) in the (blue seed) maintainer seed stream was 10 % (ranging from 7 to 14 % across the 4 genotypes), and the purity of blue aleurone seeds was at least 95 % on average across these 4 genotypes. Thus, different backlights may have a significant influence on sorting efficiency. A green colored backlight, such as any or the green backlights with HSL values as described specifically herein, may achieve best sorting results for shoot outs of blue seeds from a mixture of blue and white seeds across all genotypes tested. A red colored backlight, such as any of the red backlights with HSL values as described specifically herein (including H=0 (H=360)), may achieve highest product yields of white seeds and may best be used to separate a blue and a white fraction in a first sorting step to obtain blue seeds, before in a second or third sorting step the dark blue seeds (3n BLA seeds) are removed from the lighter blue seeds (1 n and 2n BLA seeds), wherein said second or third sorting step is best with a yellow or orange backlight color, such as the yellow or orange or red backlights with HSL values as described specifically herein. Further, a yellow or orange colored backlight, such as the yellow or orange backlight with HSL values as described specifically herein, may provide the best results for sorting out dark blue seeds having 3n BLA locus in the aleurone layer from a sample of lighter blue seeds (1 n and 2n BLA).

[0123] The method of sorting cereal seed may specifically provide better results for sorting out blue seeds compared to sorting out white seeds. The method may be used to obtain highly pure white seed lots, specifically when using at least two sorting rounds. This may be e.g. realized on standard large-scale seed sorting devices having multiple sorting rounds in the same machine using several chutes working in series. Thus, in general, for genotypes being more difficult to be sorted out, more shoot-out runs may improve the sorting purity. Multispectral imaging devices, such as by using a videometer®, may be used as a quality control system to check the purity of the sorted seeds. Additionally or alternatively, genotyping data may be used for quality check. The multispectral imaging device may generally be used for distinguishing 3n blue aleurone haplotype seeds (dark blue seeds) from “normal” blue 2n seed or light blue 1 n blue aleurone haplotype seeds. The 3n blue aleurone haplotype seeds may be significantly darker than 2n blue aleurone haplotype seeds. The amount of dark blue 3n seeds may be reduced from blue seed batches by sorting out the darkest blue seeds (see above), e.g., by reducing the Lmin (object setting) value of a recipe to 0, or to 0<Lmin<60, specifically to 0<Lmin< 50, more specifically to 0<Lmin<40, 0<Lmin<30, 0<Lmin< 20, or 0<Lmin<10, even more specifically 0<Lmin<30 (depending on the darkness of a seed lot), and the Lmax (object) value from 85 to 127 or from 81 to 123 or the Lmax value is 78<Lmax<127, depending on the percentage or amount of 3n seeds which is to be removed and depending on the darkness of a seed lot, specifically using any one of the colored backlights described herein for removing / shooting out blue aleurone seed (including those referred to in the description of Figures 4-9), such as any one of the colored backlights having a Hue value of 0, 20, 47, 71 , 340 or 111 , such as H20, H47, H71 , or H111 , preferably H20 or H71 , more specifically a yellow or orange colored backlight having a H coordinate of 20

[0124] < H < 70, an S coordinate of 0.25 < S < 1 and a L coordinate of 0.35 < L < 0.78. In particular, this backlight has a H coordinate of 20 or 71 , an S coordinate of 1 and an L coordinate of 0.58. In one embodiment, for the low L method described herein (to shoot out darker colored, such as dark blue, seeds), also a white LED backlight or a backlight with an L coordinate of L = 255 (white) may be used. In this method, the Lmin (object) value is best set at 0 (safest choice, as works well independent of darkness of seeds), but can also be raised up to 50 for dark seedlots and up to 60 for light seedlots, without significantly influencing the percentage of dark blue seeds being shot out. Also, in this (low L) method to shoot out 3n blue aleurone seeds from blue seed batches, the Lmax depends on the amount of 3n blue aleurone seeds to be removed. E.g., on a L scale from 0 to 255, the best Lmax for a 10% dark blue seed shoot is 77 < L < 87, for a

[0125] 15% shoot out is 79 < L < 95, for a 20% shoot out is 81 < L < 100, for a 30% shoot out is 84 < L

[0126] < 103, for a 40% shoot out is 87 < L < 107, and for a 50% dark blue seed shoot out is 91 < L < 110. In one embodiment, the Lmax (object) value here is set from 85 to 89, such as 85, 86, 87, 88, or 89 (on a scale from 0 to 255). The Hue object settings in this low L (or lightness) method can be from H > 345 to H < 45 or from H > 354 to H < 54, or from > 345 to H < 79, or can be any H coordinate comprising the range of H > 354 and H < 54 (on the 0-360 H scale). In one embodiment of the invention, the amount of darker seeds, such as darker cereals seeds, such as dark blue 3n cereal seeds may also be reduced from lighter seeds, such as lighter cereal seeds, such as lighter blue 1 n / 2n seeds, by sorting out the darkest seeds, such as the dark blue 3n seeds, e.g., by reducing the Lmin (object setting) value of a recipe. The Lmin can be set at 0, or to a range from 0 to 60, specifically to a range from 0 to 50, more specifically to a range from 0 to 40, 0 to 30, 0 to 20, or 0 to 10, even more specifically to a range from 0 to 30 (including the end points, depending on the darkness of a seed lot), and the Lmax (object) value from 78 to 120 or from 81 to 123 or from 78 to 127 (on a 0-255 scale, depending on the percentage or amount of 3n seeds which is to be removed, and the darkness of a seed lot).

[0127] The above (low L) method to reduce the amount of dark blue 3n seeds in blue seed batches can, e.g., be used in a second or third sorting step after first sorting out the blue seeds (retaining white seeds), using the fraction shot-out from the earlier sorting step(s) as described above (e.g., in the DB Blue Shoot Out step as schematically shown in Figures 4-9, with the adapted backlight HSL settings of the invention).

[0128] In this context, it is useful to note that the percentage of 3n (double blue, disomic) dark blue seeds in the blue seeds fraction of segregating spikes of cereal plants with a hybrid system based on a blue aleurone locus on the same chromosome (arm) as the fertility restorer gene (in a plant having a male-sterility gene, causing male sterility in absence of the restorer gene) is approximately 10-20%. If that blue seed fraction is then used for seed multiplication without depletion of the disomic dark blue seeds, the percentage of dark blue 3n seeds increases to more than 30% in the first, to more than 60% in the second and to more than 90% in the 3rd generation. Without depletion of the disomic dark blue seeds, the % of “white” seeds obtained in the amplification of the maintainer drops from about 66 / 34 (white / blue) to 60 / 40 in the next amplification, to 45 / 55 in the following amplification, and to 31 / 69 in the next amplification (provided 3 n blue aleurone seeds are not removed). Hence, the amount of male-sterile females obtained from a certain area decreases and therefore the production cost is increased. Hence, it is suggested to deplete the amount of dark blue 3n seeds in the blue seed fraction by means of a specific seed sorting step in which either the darkest seeds of a blue seed fraction are shot out, or alternatively only light blue seeds are shot out and only the light blue fraction is used for seed amplification. Said sorting step can also be used to remove (darker) colored seed from a seed batch containing (darker) colored and non-colored or light-colored seeds.

[0129] As outlined above, step ill. comprises automatically identifying, from the image taken in step ii. , seeds to be sorted out from the seed stream. In step ill., seeds to be sorted out from the seed stream may be identified by identifying, in the image, objects having predefined color coordinates as provided (on the image taken) by the sorting device, specifically color coordinates in a predefined range in the HSL color space. In one embodiment of this invention, the method may comprise the above sorting method wherein the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, and wherein the predefined object HSL settings of the seeds to be sorted out, as provided by the sorting device, may be purposefully changed or modified so as to shoot-out certain seeds or a certain amount of seeds, such as to shoot-out (a certain amount of) (darker) colored seeds from noncolored or lighter-colored seeds, or to shoot-out (a certain amount of) dark blue (3n) blue aleurone seeds from 2n / 1 n blue aleurone seeds or non-colored “white” seeds, or light blue (1 n / 2n) seeds from dark blue and non-colored seeds. In one embodiment, the modification of said predefined object HSL settings may encompass the modification of the L object settings, such as a Lmin object setting that is set lower than the Lmin object setting provided by the sorting device for the (darker) colored seeds to be shot-out, or said modification of said L object settings with a modification of the S object settings, such as a Smin object setting that is set lower than the Smin object setting provided by the sorting device for the (darker) colored seeds to be shot-out (with the Lmax or Lmax and Smax being set at a value to shoot out a certain % of (darker) colored seeds), and wherein the H object setting is any H object range, including 354 < H < 45.

[0130] In one embodiment, the amount of darker colored seeds, such as dark blue seed containing a 3n blue aleurone, may also be reduced from non-colored or lighter-colored, such as 1 n and / or 2n blue aleurone seeds, by sorting out the darkest seeds (see above), e.g., by reducing the Lmin and Smin (object setting) value of a recipe. The Lmin can be set at 0, or to a range from 0 to 60, specifically to a range from 0 to 50, more specifically to a range from 0 to 40, 0 to 30, 0 to 20, or 0 to 10, even more specifically to a range from 0 to 30 (including the end points, depending on the darkness of a seed lot), and the Lmax (object) value from 78 to 120 or from 81 to 123 or from 78 to 127 (on a 0-255 scale) depending on the percentage or amount (weight) of darker colored seeds, such as 3n blue aleurone seeds, which is to be shot-out / removed and depending on the darkness of a seed lot. The Smin can be set at 0, or to a range from 0 to 40, specifically to a range from 0 to 30, 0 to 20, 0 to 17 or 0 to 10, even more specifically to a range from 0 to 30 (including the end points, depending on the darkness of a seed lot), and the Smax (object) value from 60 to 78 or from 60 to 99 depending on the percentage or amount of darker colored seeds, such as 3n blue aleurone seeds, which is to be removed and depending on the darkness of a seed lot (on a 0-255 scale). In this low L+S method, any one of the colored backlights described herein can be used for removing / shooting out darker colored seed, such as 3n blue aleurone seed, (including those referred to in Figures 4-9), such as any one of the colored backlights having a Hue value of 0, 20, 47, 71 , 340 or 111 , such as backlights having a Hue value of 0, 20, 47, 71 , or 111 , more specifically a yellow or orange colored backlight having a H coordinate of 0 < H < 111 or 20 < H < 70, an S coordinate of 0.25 < S < 1 and a L coordinate of 0.35 < L < 0.78. In particular, this backlight has a H coordinate of 20 or 71 , an S coordinate of 1 and an L coordinate of 0.58. In one embodiment, in the low L+S method described herein (to shoot out darker colored, such as dark blue, seeds), the backlight may also be a white LED or a backlight with an L coordinate of L = 255 (white). In this method, the Lmin (object) value may be best 0, but can also be raised up to 50 for dark seedlots and up to 60 for light seedlots, without significantly influencing the percentage of dark blue seeds being shot out. Also, in this (low L+S) method to shoot out 3n blue aleurone seeds from blue seed batches, the Lmax and Smax depends on the amount of 3n blue aleurone seeds to be removed. In this method, one will typically target a certain % (by weight) of dark blue seeds to be shot-out, and by adapting the L and S settings, you get to the intended % seeds shot-out (e.g., when starting with a test batch of 1000 gram from a larger batch of mixed seeds having about 30 % dark blue seeds, and targeting a removal of 20 % of dark blue seeds, the L and S settings within the above ranges are adapted until the weight shot-out is about 200 gram - these L+S settings can then be used on the entire seed batch to remove most dark blue seeds). Generally, the low L+S method allows for a more precise adjustment to a specific amount of darker colored seeds to be shot-out compared to the low L method, as the effect of the low L+S method comes from the interplay of the L and S object settings. The Hue object settings in this low L+S method can be from H > 345 to H < 45 or from H > 354 to H < 54, or from H > 354 to H < 45, or from > 345 to H < 79, or can be any H coordinate comprising the range of H > 354 and H < 54 (on the 0-360 H scale). The above (low L+S) method to reduce the amount of dark blue 3n seeds in blue seed batches can e.g., be used in a second or third sorting step after first sorting out the blue seeds (retaining white seeds), using the fraction shot-out from the earlier sorting step(s) as described above (e.g., in the DB Blue Shoot Out step as schematically shown in Figures 4-9, with the indicated preferred backlight HSL settings provided in the description of Figures 4-9).

[0131] The % of dark (3n) blue aleurone seeds in a seed batch can be obtained by genotyping (e.g., copy number analysis of BLA locus), but also in observation trials. In observation trials, the amount of plants only producing blue spikes can be measured, and from that the average % of dark blue seeds in the next generation can be calculated (as it is expected that when grown in the same environment, a certain plant genotype will have the same or a similar male transmission rate). Once the % of dark blue seeds is estimated, one can use the low L or low L+S method as described herein to sort-out a certain % of dark blue seeds from the blue seed fraction. The percentage shoot out should not exceed the percentage 3n blue aleurone seeds (double blue seeds) estimated to be in the seedlot.

[0132] Also, the amount of dark colored seeds, such as blue aleurone 3n seeds, and non-colored “white” seeds may also be reduced from colored, such as blue, seed batches by shooting out the lighter colored, such as the lighter blue (1 n / 2n aleurone), seeds, and retaining the darkest blue and the “white” non-colored seeds (also referred to herein as the light blue (LB) shoot-out method). In this LB shoot-out method, the Lmin object setting for the light blue seed to be shot out can be: 96<Lmin^110, such as an Lmin of 105, and the Smin object setting can be: 55<Smin^75, such as 65 < Smin 75, such as an Smin of 70, while the Lmax can be: 138-145, such as an Lmax of 140, and the Smax can be: 95-255, such as an Smax of 120. In one embodiment of this LB shootout method, the Smin can be 65<Smin^75, and the Lmin can be 97<Lmin^110. In one embodiment of this light blue shoot-out method, the Smin object setting is from 55 to 62 (or 55 < Smin 62), and the Smax is 98, and the Lmin is from 96 to 102 (or 96 < Lmin 102), and the Lmax is 140 (on a 0-255 scale). This (LB shoot-out) method to reduce the amount of dark colored (such as blue 3n) and “white” seeds in colored (such as blue) seed batches can e.g., be used in a third sorting step after first sorting out the colored (such as blue) seeds (retaining white seeds) in a first sorting step, and then shooting out the (3n aleurone) dark colored seeds from said colored seeds in a second sorting step, using the fraction retained in the second sorting step, as described above (e.g., see the LB Blue Shoot Out step schematically shown in some of Figures 4-9, with the indicated preferred backlight HSL settings provided in the description of Figures 4-9).

[0133] The Hue object settings in this LB shoot-out method can be any Hue range including the range of H > 354 and H < 45, such as an H coordinate from H > 345 to H < 45 or from H > 354 to H < 54 (on the 0-360 H scale). Although the low L and certainly the low L+S methods are preferred, in some instances, this LB shoot-out method can also be used to shoot out darker colored, such as 3n aleurone dark blue seeds, from a mostly colored, such as a mostly blue aleurone, seed batch, in addition to or to replace the low L or low L+S methods above, and can also be used to shoot-out non-colored “white” seeds from a seed batch containing mostly colored seeds, such as blue aleurone seeds, and some remaining “white” seeds (be it mostly at lower efficiency than the low L or low L+S methods herein).

[0134] While it would be best to shoot-out all 3n aleurone dark blue seeds (that will only produce (fertile and blue) 3n blue aleurone progeny) from the maintainer / blue seeds in a hybrid system using blue aleurone as color marker (on the same chromosome or chromosome arm as the restorer gene that restores the male sterility used in the hybrid system), due to the overlap in color intensity of 2n and 3n Bia seeds, this could only be achieved with significant losses of maintainer seeds. In practice not all 3n blue seeds will be shot-out, as the goal is to minimize losses of maintainer seeds with 1 n / 2n aleurone that can re-produce the male-sterile female lines used as parent line in a hybrid production. The major goal with regard to the removal of the 3n aleurone seed is to optimize the efficiency of maintainer and white seed production and counteract the rapid increase of 3n blue aleurone seeds. This means that any dark blue removal shoot out should result in a higher proportion of 1 n / 2n aleurone seeds to 3n aleurone seeds.

[0135] In one embodiment of the current invention, the colored seed to be shot-out may be a (darker) colored seed where the Hue object settings (when provided with different preferred backlight colors as described herein) do not significantly change with other backlight colors (wherein a significant change is that the H object settings provided by the sorting device with one backlight color (such as H object settings 180-240 for an HO backlight) does not overlap with the H object setting provided by the sorting device with another backlight color (such as H object settings of 30-92 for an H71 backlight), or that the majority of the H object setting range provided by the sorting device for one backlight color is outside the H object setting range provided by the sorting device for another backlight color). In one embodiment of the invention, the colored seed to be shot-out may not be dark black seed.

[0136] As another example, the method may comprise to sort out only the 1 n and 2n blue aleurone seeds from a mixture of 1 n, 2n, 3n and white seeds. By shooting out the light blue seeds, the blue fraction may be depleted of white seeds and the darkest blue 3n bla seeds. The best backlight for such purpose may be a red backlight, in particular a backlight with a H coordinate of 0, a S coordinate of 1 and a L coordinate of 0.43, preferably using the modified object settings as in the above LB shoot out method. The backlight and / or the sorting settings may be individually optimized after harvest prior to performing a sorting campaign, e.g. by performing a test sorting to determine for that seed produced in that environment the best colored backlight and the best sorting settings, specifically depending on the purpose of sorting, e.g. for optimizing the amount of white non-colored seeds with no or little blue seed contaminants and / or optimizing the purity of colored seeds (like 1 n and 2n BLA seeds) with no or little non-colored white seed contaminants and / or no or little dark blue (3n BLA seeds).

[0137] Green colored background light may specifically be used for obtaining highest purity of white seeds, especially in less sorting rounds. It might be considered as the most stringent background light, but may decrease the amount of white seed compared to e.g. red colored backlight. Red colored backlight may preferably be used for obtaining high purity of above 99 %, or above 99.7 % while retaining most white seeds in the sorted seed fraction for particular seed lots. Yellow colored back light may preferably be used for particular genotypes with regard to purity and amount, and if highest purity is not required, may be the best backlight retaining most white seeds across several genotypes compared to red or green colored backlight.

[0138] As used herein, the terms “have”, “comprise” or “include” or any arbitrary grammatical variations thereof are used in a non-exclusive way. Thus, these terms may both refer to a situation in which, besides the feature introduced by these terms, no further features are present in the entity described in this context and to a situation in which one or more further features are present. As an example, the expressions “A has B”, “A comprises B” and “A includes B” may both refer to a situation in which, besides B, no other element is present in A (i.e. a situation in which A solely and exclusively consists of B) and to a situation in which, besides B, one or more further elements are present in entity A, such as element C, elements C and D or even further elements.

[0139] Further, it shall be noted that the terms “at least one”, “one or more” or similar expressions indicating that a feature or element may be present once or more than once typically are used only once when introducing the respective feature or element. In most cases, when referring to the respective feature or element, the expressions “at least one” or “one or more” are not repeated, notwithstanding the fact that the respective feature or element may be present once or more than once.

[0140] Further, as used herein, the terms "preferably", "more preferably", "particularly", "more particularly", "specifically", "more specifically" or similar terms are used in conjunction with optional features, without restricting alternative possibilities. Thus, features introduced by these terms are optional features and are not intended to restrict the scope of the claims in any way. The invention may, as the skilled person will recognize, be performed by using alternative features. Similarly, features introduced by "in an embodiment of the invention" or similar expressions are intended to be optional features, without any restriction regarding alternative embodiments of the invention, without any restrictions regarding the scope of the invention and without any restriction regarding the possibility of combining the features introduced in such way with other optional or non-optional features of the invention.

[0141] Summarizing and without excluding further possible embodiments, the following embodiments may be envisaged:

[0142] Embodiment 1 : A method of sorting seeds, wherein the seeds contain darker colored and lighter colored seeds, or non-colored seeds and colored seeds, specifically light colored seeds and / or non-colored seeds, and colored seeds, more specifically seeds containing a blue aleurone, or wherein the cereal seeds contain darker and lighter colored seeds, the method comprising: i. supplying a seed stream to a sorting station, the sorting station comprising at least one backlight device for backlighting a seed of the seed stream and at least one camera for taking at least one image of the backlighted seed;

[0143] II. taking, with the camera, at least one image of the backlighted seed of the seed stream; ill. automatically identifying, from the image taken in step ii. , seeds to be sorted out from the seed stream; and iv. automatically ejecting seeds identified to be sorted out from the seed stream, wherein the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80.

[0144] Embodiment 2: The method according to the preceding embodiment, wherein, in step ill., seeds to be sorted out from the seed stream are colored seeds identified by identifying, in the image, objects cumulatively fulfilling the following conditions: the objects have predefined color coordinates, specifically color coordinates in a predefined range in the HSL color space, and the objects have one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape.

[0145] Embodiment 3: The method according to the preceding embodiment, wherein the identification of the seeds to be sorted out comprises determining color coordinates and an area of the objects in the image, wherein the objects are determined to have the predefined color coordinates if the determined color coordinates are within a predefined range in the HSL color space; the objects are determined to have the predefined area if the determined area exceeds an area threshold.

[0146] Embodiment 4: The method according to the preceding embodiment, wherein the predefined range, in the HSL color space, has a H coordinate from 0 to 85, specifically from 5 to 65 and / or from 19 to 79, a S coordinate from 15 to 85, specifically from 18 to 82 and / or from 19 to 83, a L coordinate from 60 to 150, specifically from 91 to 141 and / or from 66 to 130, wherein the area threshold is in the range of 200 to 1500 pixels, 400 to 1500 pixels, or 500 to 1000 pixels, specifically in the range of 600 to 850 pixels, more specifically in the range of 650 to 700, most specifically is 700 pixels.

[0147] Embodiment 5: The method according to any one of the two preceding embodiments, wherein the identification of the seeds to be sorted out comprises determining at least one recognition parameter comprising weighting the color coordinates with the area of the objects in the image, specifically by using a product of the color coordinates and the area of the objects in the image.

[0148] Embodiment 6: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a H coordinate in at least one range selected from the group consisting of:

[0149] 0 < H < 25 or H > 310, specifically excluding a range of 332 < H < 338; 70 < H < 150;

[0150] 25 < H < 70.

[0151] Embodiment 7: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a S coordinate of 0.25 < S < 1 .0, specifically of 0.5 < S < 1 .0, more specifically of 0.75 < S < 1 .0.

[0152] Embodiment 8: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a S coordinate of 0.25 < S < 1 .0 and a H coordinate of 25 < H < 70.

[0153] Embodiment 9: The method according to the preceding embodiment, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.35 < L < 0.78.

[0154] Embodiment 10: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a S coordinate of 0.33 < S < 1 .0 and a H coordinate of 70 < H < 150.

[0155] Embodiment 11 : The method according to the preceding embodiment, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.28 < L < 0.63.

[0156] Embodiment 12: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a S coordinate of 0.40 < S < 1 .0 and a H coordinate of 25 < H or H > 310.

[0157] Embodiment 13: The method according to the preceding embodiment, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.13 < L < 0.78. Embodiment 14: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a H coordinate of 111 , a S coordinate of 1 .0 and a L coordinate of 0.48.

[0158] Embodiment 15: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a H coordinate of 71 , a S coordinate of 1 .0 and a L coordinate of 0.48.

[0159] Embodiment 16: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a H coordinate of 340, a S coordinate of 0.75 and a L coordinate of 0.68.

[0160] Embodiment 17: The method according to any one of the preceding embodiments, wherein the at least one backlight, in the HSL color space, has a H coordinate of 47, a S coordinate of 1 .0 and a L coordinate of 0.59.

[0161] Embodiment 18: The method according to the any one of the preceding embodiments, wherein the seeds to be sorted out from the seed stream are the seeds containing a blue aleurone.

[0162] Embodiment 19: The method according to anyone of the preceding embodiments, wherein the method is a continuous method, wherein, in step i., a continuous seed stream is supplied to the sorting station, wherein, in step ii., a continuous stream of images is taken of the seed stream, and, wherein, in step ill., the stream of images is continuously evaluated for continuously identifying seed to be sorted out from the seed stream.

[0163] Embodiment 20: The method according to any one of the preceding embodiments, wherein a batch of seeds is provided, and wherein the batch of seeds is subjected to method steps i.-iv. repeatedly, specifically at least twice, wherein, in each repetition, the batch is diminished by the seeds ejected in step iv. of the previous run.

[0164] Embodiment 21 : The method according to any one of the preceding embodiments, wherein the cereal seeds comprise, specifically are, grains of wheat.

[0165] Embodiment 22: The method according to any one of the preceding embodiments, wherein the method comprises at least one backlight identification step, specifically at least one backlight identification step preceding step i., wherein the backlight identification step comprises identifying the at least one backlight to be used for sorting.

[0166] Embodiment 23: The method according to the preceding embodiment, wherein the backlight identification step comprises performing a plurality of sortings with a plurality of backlights, each sorting comprises using an aliquot of the seeds to be sorted. Embodiment 24: The method according to the preceding embodiment, wherein the at least one backlight is determined via at least one quality control step, the quality control step comprising evaluating a purity of the sorting across the plurality of backlights, or comprising evaluating a purity of the sorting and an amount of losses of desired seeds of the sorting across the plurality of backlights.

[0167] Embodiment 25: The method according to any one of preceding embodiments, wherein, in step ill., seeds to be sorted out from the seed stream are colored seeds, such as darker colored seeds and / or lighter colored seeds.

[0168] Embodiment 26: The method according to any one of the preceding embodiments, wherein the method comprises performing steps i. to iv. with at least one first backlight, the first backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the method further comprises repeating steps i. to iv. using one of an ejected fraction of seeds or a retained fraction of seeds with at least one second backlight, the second backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the second backlight is different from the first backlight.

[0169] Embodiment 27: The method according to the preceding embodiment, wherein the first backlight is used for a first purpose of sorting cereal seeds.

[0170] Embodiment 28: The method according to the preceding embodiment, wherein the first purpose of sorting cereal seeds comprises at least one purpose selected from the group consisting of: maximize yield of non-colored cereal seeds; maximize yield of colored cereal seeds; maximize purity of non-colored cereal seeds, specifically by removing of impurities, such as weed seeds, ergot-infected seeds and / or other plant parts; maximize purity of colored cereal seeds; reduce number of colored seeds in non-colored seeds; reduce number of non-colored seeds in colored seeds; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof; sorting out of dark colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds (with or without non-colored seeds in said mixture); sorting out of lighter colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds, which mixture may contain noncolored seeds; sorting out of dark colored seeds and lighter colored seeds from a mixture of seeds containing dark colored seeds, lighter colored seeds and non-colored seeds.

[0171] Embodiment 29: The method according to any one of the two preceding embodiments, wherein the second backlight is used for a second purpose of sorting cereal seeds.

[0172] Embodiment 30: The method according to the preceding embodiment, wherein the second purpose of sorting cereal seeds comprises at least one purpose selected from the group consisting of: maximize yield of non-colored cereal seeds; maximize yield of colored cereal seeds; maximize purity of non-colored cereal seeds; maximize purity of colored cereal seeds; reduce number of colored seeds in non-colored seeds; reduce number of non-col- ored seeds in colored seeds; maximize purity of non-colored seeds while reducing losses of non-colored seed; maximize purity of colored seeds while reducing losses of colored seed; or any combinations thereof; sorting out of dark colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds (with or without non-colored seeds in said mixture); sorting out of lighter colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds, which mixture may contain noncolored seeds; sorting out of dark colored seeds and lighter colored seeds from a mixture of seeds containing dark colored seeds, lighter colored seeds and non-colored seeds.

[0173] Embodiment 31 : The method according to any one of the two preceding embodiments, wherein the second purpose of sorting cereal seeds is different from a first purpose of sorting cereal seeds.

[0174] Embodiment 32: The method according to any one of the six preceding embodiments, wherein a repetition of steps i. to iv. using one of the ejected fraction of seeds or the retained fraction of seeds comprises using a different sorting protocol compared to an initial sorting in step ill..

[0175] Embodiment 33: The method according to any one of the seven preceding embodiments, wherein, in step ill., seeds to be sorted out from the seed stream are identified by identifying, in the image, objects having color coordinates of a predefined sorting protocol, specifically color coordinates in a predefined range in the HSL color space.

[0176] Embodiment 34: The method according to the preceding embodiment, wherein the predefined sorting protocol comprises at least one protocol selected from the group consisting of: a H coordinate comprising the range of 354<H<45 or the range of H > 354 and H < 54, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax<127, and an Smax of 60<Smax^99; a H coordinate comprising the range of 354< H<45 or in the range of H > 345 and H < 79, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax^99; a H coordinate comprising the range of 354<H<45 or in the range of H > 345 and H < 45, a S coordinate in the range of Smin < S < 98, wherein Smin is in the range of 55 < Smin 62 and a L coordinate in the range of Lmin < L < 140, wherein Lmin is in the range of 96 < Lmin

[0177] < 102; a H coordinate comprising the range of 354<H<45 or in the range of H > 354 and H

[0178] < 54, a S coordinate in the range of Smin < S < 98, wherein Smin is in the range of 55 < Smin

[0179] < 62 and a L coordinate in the range of Lmin < L < 140, wherein Lmin is in the range of 96 < Lmin 102; a combined protocol with first color coordinates comprising a H coordinate comprising the range of 354<H<45 or in the range of 19 < H < 79, a S coordinate in the range of 19 < S < 83, and a L coordinate in the range of 62 < L < 130, and with second color coordinates comprising a H coordinate comprising the range of of 354<H<45 or in the range of 5 < H < 65, a S coordinate in the range of 18 < S < 82, and a L coordinate in the range of 88 < L < 145; a H coordinate comprising the range of 354<H<45, a S coordinate in the range of 35 < S < 99, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30 and wherein the Lmax is set so as to get a certain % shot-out, such as an Lmax of 78<Lmax^127; a H coordinate comprising the range of 354<H<45, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax<99; a H coordinate comprising the range of 354<H<45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 96<Lmin^110, 96<i_min^102 or 97<i_min^110, such as an Lmin of 105, and the Lmax is 138<i_max^145, such as an Lmax of 140, and an S coordinate in the range of Smin^S<Smax wherein the Smin is 55<Smin^75, 55<Smin^62 or 65<Smin<75, such as an Smin of 70, and the Smax is 95<Smax^255, such as an Smax of 98 or 120; or said predefined sorting protocol may comprise at least one protocol with HSL object settings for the seeds to be ejected / shot-out, wherein the H object setting can be any H range, as long as it includes 354 < H < 45, such as a sorting protocol with H object settings for the seeds to be ejected / shot-out of any H range, but including the range of 354 < H < 45, the Smin and Smax as provided by the sorting device for the seeds to be shot-out, and an Lmin of 0 or 0<Lmin^30 and an Lmax between 85 and 127.

[0180] Embodiment 35: The method according to any one of the two preceding embodiments, wherein the seeds to be sorted out are further identified by identifying objects in the image having one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape.

[0181] Embodiment 36: The method according to any one of the three preceding embodiments, wherein the seeds to be sorted out is further identified by identifying objects in the image having a determined area exceeding an area threshold in the range of 200 to 1500 pixels, or in the range of 400 to 1000 or 500 to 1000 pixels, specifically in the range of 550 to 800 pixels, more specifically in the range of 700 to 800, most specifically is 200, 400, 600, 700 or 750 pixels.

[0182] Embodiment 37: The method according to any one of the eleven preceding embodiments, wherein the method comprises separately repeating steps i. to iv. with the ejected fraction of seeds and the retained fraction of seeds.

[0183] Embodiment 38: The method according to any one of the twelve preceding embodiments, wherein the steps i. to iv. are repeated using the retained fraction of seeds with the first backlight, wherein the steps i. to iv. are repeated using the ejected fraction of seeds with the second backlight. Embodiment 39: The method according to the preceding embodiment, wherein the first backlight, in the HSL color space, has at least one color selected from the group consisting of: a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123.

[0184] Embodiment 40: The method according to any one of the two preceding embodiments, wherein the second backlight, in the HSL color space, has at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 0, a S coordinate of 250 and a L coordinate of 110.

[0185] Embodiment 41 : The method according to any one of the fifteen preceding embodiments, wherein the steps i. to iv. are repeated using the retained fraction of seeds with the second backlight, wherein the steps i. to iv. are repeated using the ejected fraction of seeds with a third backlight, the third backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the third backlight is different from the first and second backlight.

[0186] Embodiment 42: The method according to the preceding embodiment, wherein the first backlight, in the HSL color space, has a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110, or has a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123.

[0187] Embodiment 43: The method according to any one of the two preceding embodiments, wherein the second backlight, in the HSL color space, has at least one color selected from the group consisting of: a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123.

[0188] Embodiment 44: The method according to the preceding embodiment, wherein any of the second backlights are used to repeat at least once the steps i. to iv. using the retained fraction of seeds.

[0189] Embodiment 45: The method according to any one of the four preceding embodiments, wherein the third backlight, in the HSL color space, has at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123. Embodiment 46: The method according to any one of the twenty preceding embodiments, wherein the steps i. to iv. are repeated using the retained fraction of seeds with the second backlight, wherein the steps i. to iv. are repeated using the ejected fraction of seeds with the first backlight.

[0190] Embodiment 47: The method according to any one of the twenty-one preceding embodiments, further comprising repeating steps i. to iv. with a retained or an ejected fraction of seeds from a repetition of steps, i. to iv. using the ejected fraction of seeds, wherein the further repetition uses a fourth backlight, the fourth backlight, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the fourth backlight is different from the first and / or second backlight.

[0191] Embodiment 48: The method according to the preceding embodiment, wherein the fourth backlight, in the HSL color space, has at least one color selected from the group consisting of: a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150; a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110; a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123; a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172; a H coordinate of 111 , a S coordinate of 255 and a L coordinate of 123.

[0192] Embodiment 49: The method according to any one of the twenty-three preceding embodiments, wherein a fraction of ejected and / or retained seeds comprising, specifically consisting of, non-colored cereal seeds is added to a repetition of steps i. to iv. using the retained fraction of seeds.

[0193] Embodiment 50: The method according to any one of the twenty-four preceding embodiments, wherein the repetition of steps i. to iv. using the using the retained fraction of seeds is repeated at least twice.

[0194] Embodiment 51 : The method according to anyone of the preceding embodiments, wherein at least step ill., and optionally at least one of steps II. and iv., are at least one of computer- implemented and computer-controlled.

[0195] Embodiment 52: The method according to anyone of the preceding embodiments, wherein, in step ill., seeds to be sorted out from the seed stream are identified by a trained artificial neural network (ANN).

[0196] Embodiment 53: The method according to the preceding embodiment, further comprising at least one training step, wherein, in the training step, the ANN is trained using labeled images.

[0197] Embodiment 54: The method according to the any one of the two preceding embodiments, wherein the labeled image comprises genotyping data, wherein the genotyping data pro- vide information on seeds having a normal (non-colored) aleurone and cereal seeds containing a blue aleurone, or provide information on cereal seeds containing a 1 n, 2n or 3n blue aleurone.

[0198] Embodiment 55: The method according to any one of the preceding embodiments, wherein the method further comprises at least one second spectral seed sorting step, wherein the second spectral seed sorting step comprises determining at least one item of spectroscopic information on the seeds of the seed stream, such as near infrared spectroscopic information, wherein the at least one item of spectroscopic information is used for automatically identifying seeds to be sorted out from the seed stream.

[0199] Embodiment 56: The method according to the preceding embodiment, wherein a trained artificial neural network automatically identifies seeds to be sorted out from the seed stream based on the at least one item of spectroscopic information.

[0200] Embodiment 57: The method according to the preceding embodiment, further comprising at least one training step, wherein, in the training step, the ANN is trained using labeled spectroscopic information.

[0201] Embodiment 58: The method according to the any one of the two preceding embodiments, wherein the labeled spectroscopic information comprises genotyping data, wherein the genotyping data provide information on seeds having a normal (non-colored) aleurone and cereal seeds containing a blue aleurone, or provide information on cereal seeds containing a 1 n, 2n or 3n blue aleurone.

[0202] Embodiment 59: The method according to any one of the preceding embodiments, for sorting out colored seeds from a seed mixture containing darker and lighter colored seeds and non-colored seeds, wherein a sorting protocol uses a combination of a) the HSL object settings for the darker colored seeds as provided by the sorting device and b) the HSL object settings for the lighter colored seed as provided by the sorting device, which method can be performed once or can be repeated several times, such as repeating 1-3 times, to maximize non-colored seed purity.

[0203] Embodiment 60: The method according to any one of the preceding embodiments, wherein L object settings as provided by the sorting device for the darker colored seeds, are modified by lowering the Lmin object settings to at least the value of the darkest seed to ensure a shoot-out of the darkest seeds, and wherein the Lmax is set to a value resulting in the shoot-out of a desired % by weight of the darkest seeds, and wherein the H object setting is any H range comprising 354<H<45, such as wherein the Lmin is 0 or 0<Lmin^60 or 0<Lmin<30. Embodiment 61 : The method of any one of the preceding embodiments, wherein the colored seed is a disease-infected seed with darker seed color or darker spots compared to the non-infected seed that is lighter or has no darker spots.

[0204] Embodiment 62: A sorting device for sorting seeds, comprising:

[0205] I. at least one seed feeder for supplying a seed stream to at least one sorting station; and

[0206] II. the at least one sorting station, comprising at least one backlight device for backlighting seeds of the seed stream, the sorting station further comprising at least one camera for taking at least one image of the backlighted seed, and the sorting station further comprising at least one ejector for ejecting seeds from the seed stream, wherein the sorting device is configured for performing the method according to any one of the preceding embodiments.

[0207] Embodiment 63: The sorting device according to the preceding embodiment, further comprising at least one controller, wherein the controller is configured for performing at least step ill. of the method, and wherein, optionally, the controller further is configured for controlling at least one of steps II. and iv. of the method.

[0208] Embodiment 64: The sorting device according to any one of the preceding embodiments referring to a sorting device, further comprising at least one target chute and at least one sort- out chute, wherein the seeds ejected in step iv. are collected by the sort-out chute and wherein the remaining seed stream is collected by the target chute.

[0209] Embodiment 65: The sorting device according to any one of the preceding embodiments referring to a sorting device, wherein the sorting station comprises at least two cameras, the cameras being configured for taking images of the backlighted seed from different angles in space.

[0210] Embodiment 66: The sorting device according to the preceding embodiment, wherein the sorting station further comprises at least two backlight devices, each backlight device being assigned to a camera of the at least two cameras.

[0211] Embodiment 67: The sorting device according to any one of the two preceding embodiments, wherein the cameras are configured for taking images of the backlighted seed from opposing directions.

[0212] Embodiment 68: The sorting device according to any one of the preceding embodiments referring to a sorting device, wherein the sorting station is configured such that the seed stream passes the sorting station between the backlight device and the camera.

[0213] Embodiment 69: The sorting device according to any one of the preceding embodiments referring to a sorting device, wherein the ejector comprises at least one of a mechanical ejector having at least one mechanical actor and a pneumatic ejector having at least one nozzle for directing an air jet at the seed to be sorted out from the seed stream.

[0214] Embodiment 70: A computer program comprising instructions which, when the program is executed by the sorting device according to any one of the preceding embodiments referring to a sorting device, specifically by the controller of the sorting device, causes the sorting device to perform at least step ill. of the method according to any one of the preceding embodiments referring to a method, and, optionally, at least one of steps II. and iv. of the method.

[0215] Embodiment 71 : A computer-readable storage medium, specifically a non-transient computer readable medium, comprising instructions which, when the instructions are executed by the sorting device according to any one of the preceding embodiments referring to a sorting device, specifically by the controller of the sorting device, cause the sorting device to perform at least step ill. of the method according to any one of the preceding embodiments referring to a method, and, optionally, at least one of steps II. and iv. of the method.

[0216] Embodiment 72: A use of the sorting device according to any one of the preceding embodiments referring to a sorting device for a purpose of use, selected from the group consisting of: sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds; sorting out of dark colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds (with or without non-colored seeds in said mixture); sorting out of lighter colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds, which mixture may contain non-colored seeds; sorting out of dark colored seeds and lighter colored seeds from a mixture of seeds containing dark colored seeds, lighter colored seeds and non-colored seeds; sorting out of colored cereal seeds from a mixture of cereal seeds containing non-colored cereal seeds and colored cereal seeds; sorting out disease-infected seeds with darker seed color from noninfected seed; sorting out of colored cereal seeds containing a blue aleurone from a mixture of cereal seeds containing non-colored cereal seeds and cereal seeds containing a blue aleurone; sorting out of dark blue aleurone seeds from a mixture of seeds containing dark blue aleurone seeds and light blue aleurone seeds (with or without non-colored seeds in said mixture); sorting out of light blue aleurone seeds from a mixture of seeds containing dark blue aleurone seeds and light blue aleurone seeds (with or without noncolored seeds in said mixture); sorting out of dark blue aleurone seeds and light blue aleurone seeds from a mixture of seeds containing dark blue aleurone seeds, light blue aleurone seeds and non-colored seeds; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 70 < H < 150; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 25 < H < 70; sorting out of col- ored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 0 < H < 25 or H > 310.

[0217] Embodiment 73: The method according to any one of the preceding embodiments referring to a method, for sorting out the dark and lighter colored seeds from a seed mixture containing dark and lighter colored seeds and non-colored seeds, such as a mixture of blue aleurone seeds and non-colored seeds, wherein the sorting protocol uses a combination of the HSL object settings for the dark colored seed and the HSL object settings for the lighter colored seed as can be provided by the sorting device, such as by sorting out dark and light blue aleurone seeds from a mixture containing dark and light blue aleurone seeds and non-colored seeds, by sorting out : (1) objects in the image having a H coordinate of 5<H<65, a S coordinate of 18<S82, an L coordinate of 88<L<145 and a size of 200-1500, or 400- 1000, or at least 700 pixels and (2) objects in the image having a H coordinate of 19<H<79, a S coordinate of 19<S<83 and a L coordinate of 62<L<130 and a size of 200-1500, or 400-1000, or at least 700 pixels, or said objects in the image have a H coordinate in (1 ) and (2) comprising the range of 354<H<45, preferably using an HSL backlight setting with a Hue value of 340, 71 or 0, which method can be performed once or can be repeated several times, such as repeating 1-3 times, to maximize non-colored seed purity.

[0218] Embodiment 74: The method according to any one of the preceding embodiments referring to a method, wherein the object settings as provided by the sorting device for the colored seeds to be sorted-out, such as dark blue aleurone seeds or light blue aleurone seeds, are modified by adapting the L object settings or by adapting the S and L object settings to increase the amount of colored seeds sorted out, and wherein the H object setting is any H range comprising 354<H<45.

[0219] Embodiment 75: The method according to any one of the preceding embodiments referring to a method, wherein the amount of dark colored seed in a batch of dark and light colored seed, such as the amount of dark blue seed containing a 3n blue aleurone in a batch of blue aleurone seeds, is reduced by reducing the Lmin object setting values of a recipe, such as an Lmin of 0 or 0<Lmin^60 or 0<Lmin^30, wherein the Lmax is set so as to get a certain % or amount of dark colored seed shot-out, such as with an Lmax of 78<Lmax^127, depending on the % or amount of dark colored seed, such as 3n blue aleurone seed, which is to be removed and depending on the darkness of a seed lot, and such a method wherein the backlight is as described in any of the preceding embodiment, or is a white LED backlight or a backlight with an L coordinate of L = 255 (white).

[0220] Embodiment 76: The method according to the preceding embodiment, wherein the Lmax depends on the amount of dark colored seed, such as dark blue seed containing a 3n blue aleurone, to be removed, and wherein the Lmax for a 10% dark colored seed shoot-out is 77 < L < 87, for a 15% dark colored seed shoot-out is 79 < L < 95, for a 20% dark colored seed shoot-out is 81 < L < 100, for a 30% dark colored shoot-out is 84 < L < 103, for a 40% dark colored shoot-out is 87 < L < 107, and for a 50% dark colored seed shoot-out is 91 < L < 110.

[0221] Embodiment 77: The method according to any one of the preceding embodiments referring to a method, wherein the amount of dark colored seed, such as dark blue seed containing a 3n blue aleurone in a batch of blue aleurone seeds, is reduced by reducing the Lmin and Smin object setting values of a recipe, such as an Lmin of 0, or 0<Lmin^60, or 0<Lmin^30, and an Smin Of 0, Or O^Smin— 40, Or O^Smin— 30, SUCh aS With an Lmax Of 78^Lmax— 120 or 81 <Lmax^123, or 78<Lmax<127, and / or an Smax of 60<Smax^78 or 60<Smax^ 99 depending on the % or amount of dark seeds, such as 3n blue aleurone seeds, which is to be removed and depending on the darkness of a seed lot, and such a method wherein the backlight is as described in any of the preceding embodiment, or is a white LED backlight or a backlight with an L coordinate of L = 255 (white).

[0222] Embodiment 78: The method according to any one of the preceding embodiments referring to a method, wherein the amount of dark colored seed, such as dark blue seed containing a 3n blue aleurone in a batch of blue aleurone seeds, is reduced by shooting out the lighter colored seeds, such as the light blue 1 n or 2n blue aleurone seeds, and retaining the darkest colored and the non-colored seeds, by using the HSL backlights in a sorting protocol with the following object settings : an Lmin of 96<Lmin^110, an Smin of 55<Smin^75, such as 65 < Smin — 75, such as wherein the Lmax is 1 38^Lmax— 145, and the Smax is 95<S max <255.

[0223] Embodiment 79: The method according to any one of the preceding embodiments referring to a method, wherein the sorting method is combined with, preceded by or followed by another sorting method using visual and / or infrared spectral analysis, specifically near-infrared spectral analysis, and / or UV spectral analysis, and / or spectral analysis using x-rays, and / or spectral analysis using Raman scattering, with or without using a trained neural network to improve sorting efficiency, e.g. to shoot out non-colored seeds from a mixture of colored and non-colored seeds.

[0224] Embodiment 80: The method or use according to any one of the preceding embodiments referring to a method, wherein with the HSL coordinates for the at least one backlight, such as either yellow, orange, red and / or green backlight, the purity of the non-colored seed that is retained, such as when sorting a mixture of blue aleurone seed and non-colored seed, is at least 99 %, specifically at least 99.7 %, more specifically that purity is obtained by repeated shoot-outs, in particular when repeating the sorting for two or three times, such as wherein said purity can be determined by genotyping the seeds for absence of the blue aleurone locus.

[0225] Embodiment 81 : The method or use according to the preceding embodiment, wherein the amount of losses of the non-colored seed that is ejected during the sorting to achieve that purity, such as when sorting a mixture of blue aleurone seed and non-colored seed, is less than 35 %, less than 30 %, less than 25 %, less than 20 %, less than 15 %, less than 10%, or less than 5 %.

[0226] Embodiment 82: The method according to any one of the preceding embodiments referring to a method, wherein with the HSL coordinates for the at least one backlight, such as either yellow, orange, red and / or green backlight, the % non-colored seeds in the colored seed, such as the blue aleurone seed, that is obtained, is less than 10 %, specifically less than 5 %, more specifically that purity is obtained when repeating the sorting, in particular when repeating the sorting for two or three times.

[0227] Embodiment 83: The method according to any one of the preceding embodiments referring to a method, wherein any given HSL coordinate for the at least one backlight, in the HSL color space, includes any given H coordinate + / - 3 (in the 0-360 H scale), includes any given S coordinate + / - 15 (in the 0-255 S scale), and / or includes any given L coordinate + / - 5 (in the 0-255 L scale).

[0228] Embodiment 84: The method according to any one of the preceding embodiments referring to a method, wherein the seeds to be sorted comprise, specifically are, cereal seeds, specifically lighter colored and / or non-colored cereal seeds and (dark and / or light) colored cereal seeds, more specifically lighter colored and / or non-colored cereal seeds and darker-colored cereal seeds.

[0229] Embodiment 85: The method according to any one of the preceding embodiments referring to a method, wherein the seeds contain lighter colored seeds and darker colored seeds, or non-colored seeds and colored seeds.

[0230] Short description of the Figures

[0231] Further optional features and embodiments will be disclosed in more detail in the subsequent description of embodiments, preferably in conjunction with the dependent claims. Therein, the respective optional features may be realized in an isolated fashion as well as in any arbitrary feasible combination, as the skilled person will realize. The scope of the invention is not restricted by the preferred embodiments. The embodiments are schematically depicted in the Figures. Therein, identical reference numbers in these Figures refer to identical or functionally comparable elements.

[0232] In the Figures:

[0233] Figure 1 shows an embodiment of a sorting device for sorting seeds in a side cut view;

[0234] Figure 2 shows a flow chart of an embodiment of a method of sorting seeds; Figures 3A and 3B show exemplary results of the method of sorting cereal seeds;

[0235] Figures 4 to 9 show different embodiments of a method of sorting cereal seeds; and

[0236] Figure 10 shows a graphical representation of the average distribution of the 3 different fractions of blue seeds within a blue fraction of segregating spikes according to genotypic and color sorting results.

[0237] Detailed description of the embodiments

[0238] Figure 1 shows an exemplary embodiment of a sorting device 110 for sorting seeds 112 in a side cut view. In this example, the seeds may comprise cereal seeds 112 containing non-col- ored cereal seeds 114 and seeds containing a blue aleurone 116. For example, the cereal seeds 112 may be wheat seeds. However, other seeds may also be feasible.

[0239] The sorting device 110 comprises at least one seed feeder 118 for supplying a seed stream 120 to at least one sorting station 122. As shown in Figure 1 , in this exemplary embodiment, the seed feeder 118 may comprise at least one feed hopper 124 configured for receiving a plurality of seeds 112 and for providing the seeds 112 to at least one chute 126 in a controllable fashion. The seed feeder 118 may further comprise at least one vibratory feeder 128 configured applying vibrations to the feed hopper 124 such that seeds 112 comprised therein may leave the feed hopper 124 to the at least one chute 126.

[0240] Further, the sorting device 110 comprises the at least one sorting station 122. The sorting station 122 comprises at least one backlight device 130 for backlighting seeds 112 of the seed stream 120. The sorting station further comprises at least one camera 132 for taking at least one image of the backlighted seed. As can be seen in Figure 1 , the sorting station 122 may be configured such that the seed stream 120 passes the sorting station 122 between the backlight device 130 and the camera 130. Further, as shown in Figure 1 , the sorting station 122 may comprise at least one front light device 131 for illuminating the seed stream 120 on a front side. The sorting station may comprise two, four or even more front light devices 131. The front light device 131 may comprise a white LED.

[0241] Further, in the exemplary embodiment of Figure 1 , the sorting station 122 may comprise at least two cameras 132. The cameras 132 may be configured for taking images of the backlighted seed from different angles in space. The sorting station 122 may further comprise at least two backlight devices 130. Each backlight device 130 may be assigned to a camera 132 of the at least two cameras 132. The cameras 132 may be configured for taking images of the backlighted seed from opposing directions. For example, the sorting station 122 may comprise at least one first backlight device 134 assigned to at least one first camera 136. The first backlight device 134 may be arranged on a first side 138 with respect to the seed stream 120. The first camera 136 may be arranged on a second side 140 opposing the first side 138 with respect to the seed stream 120. Further, the sorting station 122 may comprise at least one second backlight device 142 and at least one second camera 144. The second backlight device 142 may be arranged on the second side 140 and the second camera 144 may be arranged on the first side 138.

[0242] The sorting station 122 further comprises at least one ejector 146 for ejecting seeds 112 from the seed stream 120. The ejector may comprise a pneumatic ejector 148 having at least one nozzle for directing an air jet at the seed to be sorted out from the seed stream 120. For example, the ejector 146 may comprise the at least one pneumatic ejector 148 configured for ejecting the seed to be sorted out from the seed stream 120 by using compressed air directed via nozzles to separate the seed to be sorted out from the seed stream 120.

[0243] The sorting device 110 may further comprise at least one target chute 150 and at least one sort- out chute 152. In this example, the seeds to be sorted out may comprise the seeds containing a blue aleurone 116. The seeds ejected by the ejector 146 may be collected by the sort-out chute 152. The remaining seed stream may be collected by the target chute 150. The sorting station 122, specifically the ejector 146, may be configured for separating the seeds 112 from the seed stream into the sort-out chute 152 in case the seeds 112 are identified as seeds to be sorted out. The sorting device 110 may be configured such that seeds 112 from the seed stream 120 which are not identified as seeds to be sorted out may pass the sorting station 122 towards the target chute 150.

[0244] The sorting device 110 may further comprise at least one controller 154. As shown in Figure, 1 , the controller 154 may be configured for communicating, e.g. via wireless and / or wired means, with other devices of the sorting device 110, e.g. with the seed feeder 118 and / or with the sorting station 122.

[0245] The sorting device 110 is configured for performing the method according to the present invention, such as according to the exemplary embodiment of Figure 2 and / or according to any other embodiment disclosed herein. Thus, for a description of the method, reference is made to the description of Figure 2.

[0246] Figure 2 shows a flow chart of an embodiment of a method of sorting seeds 112. In this example, the seeds may comprise cereal seeds 112 containing non-colored cereal seeds 114 and colored seeds, specifically seeds containing a blue aleurone 116. In this exemplary embodiment, the seeds to be sorted out from the seed stream 120 may specifically be the seeds containing a blue aleurone 116. However, other seeds may also be feasible. The method may further comprise using a sorting device 110, e.g. the exemplary embodiment of the sorting device 110 shown in Figure 1 .

[0247] The method comprises the following steps that may be performed in the given order. However, a different order may also be possible. In particular, one, more than one or even all of the method steps may be performed once or repeatedly. Further, the method steps may be performed successively or, alternatively, one or more of the method steps may be performed in a timely overlapping fashion or even in a parallel fashion and / or in a combined fashion. The method may further comprise additional method steps that are not listed.

[0248] The method comprises: i. (denoted by reference number 156) supplying the seed stream 120 to the sorting station 122, the sorting station 122 comprising the at least one backlight device 130 for backlighting a seed 112 of the seed stream 120 and the at least one camera 132 for taking at least one image of the backlighted seed;

[0249] II. (denoted by reference number 158) taking, with the camera 132, at least one image of the backlighted seed of the seed stream 120; ill. (denoted by reference number 160) automatically identifying, from the image taken in step ii., seeds to be sorted out from the seed stream 120; and iv. (denoted by reference number 162) automatically ejecting seeds identified to be sorted out from the seed stream 120, wherein the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80.

[0250] The at least one backlight, in the HSL color space, may specifically have a H coordinate in at least one range selected from the group consisting of:

[0251] 0 < H < 25 or H > 310, specifically excluding a range of 332 < H < 338;

[0252] 70 < H < 150;

[0253] 25 < H < 70.

[0254] Specifically, the H coordinate in the range 0 < H < 25 or H > 310 may comprise red backlight. The H coordinate in the range 70 < H < 150 may comprise green backlight. The H coordinate in the range 25 < H < 70 may comprise yellow backlight. The at least one backlight, in the HSL color space, may have a S coordinate of 0.25 < S < 1 .0, specifically of 0.5 < S < 1 .0, more specifically of 0.75 < S < 1 .0.

[0255] In step ill., seeds to be sorted out from the seed stream 120 may be colored seeds identified by identifying, in the image, objects cumulatively fulfilling the following conditions: the objects have predefined color coordinates, specifically color coordinates in a predefined range in the HSL color space, and the objects have one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape.

[0256] Specifically, the identification of the seeds to be sorted out may comprise determining color coordinates and an area of the objects in the image, wherein the objects are determined to have the predefined color coordinates if the determined color coordinates are within a predefined range in the HSL color space; the objects are determined to have the predefined area if the determined area exceeds an area threshold.

[0257] Further, the method may specifically be a continuous method. Specifically, the continuous method of sorting cereal seeds 112 may comprise performing the methods steps repeatedly and at least partially overlapping in time. In step i., a continuous seed stream may be supplied to the sorting station 122. In step ii., a continuous stream of images may be taken of the seed stream 120. In step ill., the stream of images may be continuously evaluated for continuously identifying seed to be sorted out from the seed stream 120.

[0258] For example, in the method, a batch of seeds may be provided. The batch of seeds may be subjected to method steps i.-iv. repeatedly, specifically at least twice. In each repetition, the batch may be diminished by the seeds ejected in step iv. of the previous run. Thus, by repeating performing method steps i.-iv. on the batch of seeds, the purity of seeds passing the sorting station may be enhanced.

[0259] Figures 3A and 3B show exemplary results of the method of sorting cereal seeds 112. Specifically, Figures 3A and 3B show the exemplary results of the method of sorting cereal seeds 112 as obtained in Example 1 , which will be described in further detail below. For a detailed description of this example, reference is made to the description of Example 1 .

[0260] Figure 3A shows a polar plot of the HSL color space, wherein the H coordinate is shown as the angle 164 around the center of the polar plot with H e [0, 360) and the L coordinate is shown in relative terms as the distance from the center with L e [0,1], In Figure 3A, the H and S coordinates of the backlight which were found in Example 1 to achieve highest relative purity in sorting out seeds containing a blue aleurone 116 are highlighted by boxes 168. As can be seen in Figure 3A, backlight having a H coordinate in between the borders at 1-1=150 and at 1-1=310 may be suitable backlight, excluding for a range of 332 < H < 338. Best performing backlight colors may specifically comprise at least one of the following: a yellow backlight having a H coordinate of 47, a S coordinate of 1 .0 and a L coordinate of 0.59; a green backlight having a H coordinate of 111 , a S coordinate of 1 .0 and a L coordinate of 0.48; a green backlight having a H coordinate of 71 , a S coordinate of 1 .0 and a L coordinate of 0.48; a red backlight having a H coordinate of 340, a S coordinate of 0.75 and a L coordinate of 0.68. These suitable backlight intervals are shown in Figure 3B in more detail.

[0261] Specifically, Figure 3B shows a diagram of the highlighted backlight colors of Figures 3A. The diagram of Figure 3A shows the L coordinate of the backlight on the x axis 170, both in relative and absolute terms, and the H coordinate of the backlight on the y axis 172. Additionally, in the diagram of Figure 3B, the RBG values corresponding to the respective HSL values are indicated on the y axis 172.

[0262] Figures 4 to 9 show different embodiments of a method of sorting cereal seeds. Therein, a mixed fraction of cereal seeds 200 comprising, specifically consisting of, non-colored cereal seeds 114 and seeds containing a blue aleurone 116 may be sorted. The embodiments shown in Figures 4 to 9 may specifically comprise performing steps i. to iv. with at least one first backlight 202, the first backlight 202, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, and further repeating steps i. to iv. using one of an ejected fraction of seeds 204 or a retained fraction of seeds 206 with at least one second backlight 208, the second backlight 208, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the second backlight 208 is different from the first backlight 202. In Figures 4 to 9, each box with crossing arrows represents a single round of sorting by performing steps i. to iv. with the specific backlight color, wherein the thick arrow represents a stream of the retained fraction of seeds 206 and the thin arrow represents a stream of the ejected fraction of seeds 208. In Figs. 4-9, the blue aleurone seed 116 can be 3n blue aleurone seed (dark blue color, disomic embryo) as indicated by the drawing of a seed with horizontal bars, or can be 2n blue aleurone seed (blue color, monosomic) as indicated by the drawing of a seed with black dots, or can be 1 n blue aleurone seed (light blue, monosomic) as indicated by the drawing of a seed with sloping bars (non-colored seed 114 is indicated by the drawing of a seed with no bars or dots (white surface)).

[0263] In the embodiments of Figures 4 to 7, the steps i. to iv. may be repeated using the retained fraction of seeds 206 with the first backlight 202, wherein the steps i. to iv. may be repeated using the ejected fraction of seeds 204 with the second backlight 208. In Figs. 4-9, the first sorting step and the further sorting steps in the triangle at the left of each Figure represent the sorting steps in the non-colored (such as “white” male-sterile seed lacking a blue aleurone) seed stream (where the intent is to get purer non-colored seeds (reducing / removing colored, such as blue aleurone, seeds)), and the first sorting step and the further sorting steps in the triangle at the right of each Figure represent the sorting steps in the colored (such as fertile seed having a blue aleurone) seed stream (where the intent is to get purer colored seeds (reducing / removing non-colored, and dark blue 3n blue aleurone seeds)). The numbers 1 , 2, 3 and 4 in a circle in Figs. 4-9 refer to the sorting rounds.

[0264] Figure 4 shows a first embodiment of a method of sorting cereal seeds with repeating steps i. to iv.. In this exemplary embodiment, a first round of sorting 210 may be performed on the mixed fraction of cereal seeds 200 with the first backlight 202 having, in the HSL color space, a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172. For this first round of sorting 210, in step ill., seeds to be sorted out from the seed stream may be identified by identifying, in the image, objects having color coordinates of a predefined sorting protocol, specifically color coordinates in a predefined range in the HSL color space. For example, the predefined sorting protocol may comprise a combined protocol with first color coordinates comprising a H coordinate in the range of 19 < H < 79, a S coordinate in the range of 19 < S < 83, and a L coordinate in the range of 62 < L < 130, and with second color coordinates comprising a H coordinate in the range of 5 < H < 65, a S coordinate in the range of 18 < S < 82, and a L coordinate in the range of 88 < L < 145. As can be seen in Figure 4, the method may comprise repeating steps i. to. iv. at least once, preferably twice, more preferably three times, with the retained fraction of seeds 206 and the first backlight 202 and the combined sorting protocol in a second round of sorting 212 to obtain a fraction of sterile white seeds of high purity of above 99 %. The ejected fraction of seeds 204 in the second round of sorting 212 may be a waste fraction of cereal seeds 214.

[0265] Further, the embodiment of Figure 4 may further comprise a third round of sorting 216 with the ejected fraction of seeds 204 from the first round of sorting 210. The third round of sorting 216 may comprise using the second backlight 208 having, in the HSL color space, a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150. Further, for this third round of sorting 216, in step ill., seeds to be sorted out from the seed stream may be identified by identifying, in the image, objects having color coordinates of a predefined sorting protocol, specifically color coordinates in a predefined range in the HSL color space. For example, the predefined sorting protocol may comprise a H coordinate in the range of H > 354 and H < 45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax is set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127 (the S coordinate is as provided by the sorting device for the dark colored seeds to be sorted out), or may comprise a H coordinate in the range of H > 354 and H < 45, an S coordinate in the range wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<LmaLmin is 0 or 0<Lmin^30, and wherein the Lmaxand Smax are set so as to get a certai ored seeds shot-out, such as an Lmaxof 78<Lmax^127, and an Smax of 60<Smax^99 fraction of seeds 204 of the third round of sorting 216 may be part of the waste fraction of cereal seeds 214. The retained fraction of seeds 206 of the third round of sorting 216 may be used for a further, fourth round of sorting 218, wherein, for the fourth round of sorting 218 the second backlight 208 and a further predefined sorting protocol may be used. In the fourth round of sorting 218, the predefined sorting protocol may have, in the HSL space, a H coordinate comprising the range of H > 354 and H < 45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 96<Lmin^110, such as an Lmin of 105, and the Lmaxis 138<Lmax^145, such as an Lmaxof 140, and an S coordinate in the range of Smin^S<Smax wherein the Smin is 55<Smin^75, such as 65 < Smin 75, such as an Smin of 70, and the Smax is 95<Smax^255, such as an Smax of 98 or 120. The retained fraction of cereal seeds 206 of the fourth round of sorting 218 may be part of the waste fraction of cereal seeds 214. The ejected fraction of seeds 204 of the fourth round of sorting 218 may provide a maintainer fraction of cereal seeds with a high purity of above 95 % of seeds containing a blue aleurone 116.

[0266] Figure 5 shows a second embodiment of a method of sorting cereal seeds with repeating steps i. to iv.. The embodiment of Figure 5 widely corresponds to the embodiment of Figure 4. Thus, for a detailed description thereof, reference is made to the description of Figure 4. However, in this exemplary embodiment, the first backlight 202 in the first round of sorting 210 and in the second round of sorting 212 may have, in the HSL color space, a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123.

[0267] Figures 6 and 7 show a third and a fourth embodiment of a method of sorting cereal seeds with repeating steps i. to iv.. The embodiments of Figures 6 and 7 widely correspond to the embodiment of Figures 4 and 5, respectively. Thus, for a detailed description thereof, reference is made to the description of Figures 4 and 5: In the third embodiment of Figure 6, the first backlight 202 in the first round of sorting 210 and in the second round of sorting 212 may have, in the HSL color space, a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172 similar to the embodiment of Figure 4. In the fourth embodiment of Figure 7, the first backlight 202 in the first round of sorting 210 and in the second round of sorting 212 may have, in the HSL color space, a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123 similar to the embodiment of Figure 5. In both embodiments, the second backlight 208 used in the third round of sorting 216 using the ejected fraction of seeds 204 from the first round of sorting 210 may have, in the HSL color space, a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150.

[0268] However, in the embodiments of Figures 6 and 7, the method may comprise repeating steps i. to iv. with a retained fraction of seeds 206 from a repetition of steps, i. to iv. using the ejected fraction of seeds 204, wherein the further repetition may use a fourth backlight 220 the fourth backlight 220, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The fourth backlight 220 may specifically be different from the first 202 and / or second backlight 208. In these exemplary embodiments, the fourth backlight 220 may be used in the fourth round of sorting 218. The fourth backlight 220, in the HSL color space, may have a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110, or a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172. The predefined sorting protocol in this fourth round of sorting 218 may e.g. comprise the above-identified combined protocol.

[0269] Further, as can be seen in Figures 6 and 7, the fraction of retained seeds 206 comprising, specifically consisting of, non-colored cereal seeds 114 may be added to a repetition of steps i. to iv. using the retained fraction of seeds 206. Specifically, in these examples, the fraction of retained seeds 206 of the fourth round of sorting 218 comprising, specifically consisting of, noncolored cereal seeds 114 may be added to the repetition of steps i. to iv. using the retained fraction of seeds 206 of the first round of sorting 210. In other words, the fraction of retained seeds 206 of the fourth round of sorting 218 may be added to the second round of sorting 212.

[0270] Figures 8 shows a fifth embodiment of a method of sorting cereal seeds with repeating steps i. to iv.. In this exemplary embodiment, the steps i. to iv. may be repeated using the retained fraction of seeds 206 with the second backlight 208, wherein the steps i. to iv. may be repeated using the ejected fraction of seeds 204 with a third backlight 222, the third backlight 222, in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80. The third backlight 202 may specifically be different from the first 202 and second backlight 208.

[0271] For example, the first backlight 202 used in the first round of sorting 210 may have, in the HSL color space, a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110. The predefined sorting protocol used in step ill. of this first round of sorting 210 may e.g. comprise the above-identified combined protocol. The second round of sorting 212 may be performed using the retained fraction of seeds 206 from the first round of sorting 210. The second backlight 208 used in the second round of sorting 212 may have, in the HSL color space, a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172. The same combined protocol may be used as the predefined sorting protocol in step ill. of the second round of sorting 212. The second backlight 208 may be used to repeat at least once the steps i. to iv. using the retained fraction of seeds 208 from the first round of sorting 210, preferably at least once, more preferably twice, or even more preferably three times. The ejected fraction of seeds 204 in the second round of sorting 212 may be part of the waste fraction of cereal seeds 214.

[0272] Further, as can be seen in Figure 8, the third round of sorting 216 may be performed using the ejected fraction of seeds 204 from the first round of sorting 210. In this third round of sorting 216, the third backlight 222, in the HSL color space, may have a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150. Further, for this third round of sorting 216, in step ill., the predefined sorting protocol may comprise a H coordinate in the range of H > 354 and H < 45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax is set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127 (the S coordinate is as provided by the sorting device for the seeds to be sorted out), or may comprise a H coordinate in the range of H > 354 and H < 45, an S coordinate in the range of Smin— S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin— L^Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127, and an Smax of 60<Smax^99. The ejected fraction of seeds 204 of the third round of sorting 216 may be part of the waste fraction of cereal seeds 214. The retained fraction of seeds 206 of the third round of sorting 216 may be used for a further, fourth round of sorting 218. A fourth backlight 220 may be used for the fourth round of sorting 218; e.g., a H coordinate of 71 , a S coordinate of 255 and a L coordinate of 123. In the fourth round of sorting 218, the predefined sorting protocol may have, in the HSL space, a H coordinate in the range of H > 345 and H < 45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 96<Lmin^110, such as an Lmin of 105, and the Lmaxis 138<Lmax^145, such as an Lmaxof 140, and an S coordinate in the range of Smin^S<Smax wherein the Smin is 55<Smin<75, such as 65 < Smin 75, such as an Smin of 70, and the Smax is 95<Smax^255, such as an Smax of 98 or 120.

[0273] As shown in Figure 8, this exemplary embodiment may also be used to obtain a fraction of sterile white seeds of high purity of above 99 % via the second round of sorting 212 and a maintainer fraction of cereal seeds with a high purity of above 95 % of seeds containing a blue aleurone 116, via the fourth round of sorting 218.

[0274] Figure 9 shows a sixth embodiment of a method of sorting cereal seeds with repeating steps i. to iv.. In this exemplary embodiment, the steps i. to iv. may be repeated using the retained fraction of seeds 206 with the second backlight 208 and, further, the steps i. to iv. may also be repeated using the ejected fraction of seeds 204 with the first backlight 202. For example, the first backlight 202 used in the first round of sorting 210 may have, in the HSL color space, a H coordinate of 0, a S coordinate of 255 and a L coordinate of 110. The predefined sorting protocol used in step ill. of this first round of sorting 210 may e.g. comprise the above-identified combined protocol. The second round of sorting 212 may be performed using the retained fraction of seeds 206 from the first round of sorting 210. The second backlight 208 used in the second round of sorting 212 may have, in the HSL color space, a H coordinate of 340, a S coordinate of 192 and a L coordinate of 172. The same combined protocol may be used as the predefined sorting protocol in step ill. of the second round of sorting 212. The second backlight 208 may be used to repeat at least once the steps i. to iv. using the retained fraction of seeds 208 from the first round of sorting 210, preferably at least once, more preferably twice, or even more preferably three times. The ejected fraction of seeds 204 in the second round of sorting 212 may be part of the waste fraction of cereal seeds 214.

[0275] Further, as can be seen in Figure 9, the first backlight 202 and the combined sorting protocol may also be used for the repetition of steps i. to iv. in the third round of sorting 216 with the ejected fraction of seeds 204 from the first round of sorting 210. The ejected fraction of seeds 204 of the third round of sorting 216 may be used for the further, fourth round of sorting 218, whereas the retained fraction of seeds 206 of the third round of sorting 216 comprising, specifically consisting of, non-colored cereal seeds 114, may be added to the second round of sorting 212. Further, in the fourth round of sorting 218, the fourth backlight 220 may be used having, in the HSL color space, a H coordinate of 20, a S coordinate of 255 and a L coordinate of 150. For this fourth round of sorting 218, in step ill., the predefined sorting protocol may comprise a H coordinate in the range of H > 354 and H < 45, a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax is set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax^127 (the S coordinate is as provided by the sorting device for the seeds to be sorted out), or may comprise a H coordinate in the range of H > 354 and H < 45, an S coordinate in the range of Smin^S<Smax wherein the Smin is 0 or 0<Smin^30, and a L coordinate in the range of Lmin^L<Lmax wherein the Lmin is 0 or 0<Lmin^30, and wherein the Lmax and Smax are set so as to get a certain % of dark colored seeds shot-out, such as an Lmax of 78<Lmax<127, and an Smax of 60<Smax^99. The ejected fraction of seeds 206 from the fourth round of sorting 218 may be part of the waste fraction of cereal seeds 214.

[0276] As shown in Figure 9, this exemplary embodiment may also be used to obtain a fraction of sterile white seeds of high purity of above 99 % via the second round of sorting 212 and a maintainer fraction of cereal seeds with a high purity of above 95 % of seeds containing a blue aleurone 116 via the fourth round of sorting 218.

[0277] Also, in the exemplary embodiments shown in Figures 4 to 9, any other sorting method can be used to shoot-out any remaining non-colored seeds (lacking a blue aleurone) before, during or after the sorting steps done in the maintenance breeding seed sorting stream, using visual and / or infrared spectral analysis, specifically near-infrared spectral analysis, and / or UV spectral analysis, and / or spectral analysis using x-rays, and / or spectral analysis using Raman scattering, and / or multispec analysis, with or without using a trained neural network to improve shooting- out of non-colored seeds, which other sorting method can be done on another or the same sorting device as the method of the invention (e.g., the Cimbria SEA.IQ PLUS or an H series sorter from AnySort, with or without adaptable backlight color).

[0278] Figure 10 shows a graphical representation of the average distribution of the 3 different fractions of blue seeds within a blue fraction of segregating spikes according to genotypic and color sorting results. The bar underneath indicates the lightness scale from light (255) to dark (0), and the approximate location of an Lmax of 89, 87 and 85 (with the 5% , 10% and 20 % indicated being the shoot-out target in percent weight). L: Light, D: Dark, X-axis: relative lightness (RL), Y- axis: relative amount of different seed types with 1 n blue aleurone, 2n blue aleurone, and 3n blue aleurone (RA).

[0279] The present invention is further illustrated by the following Examples.

[0280] Example 1

[0281] In the first example, the method was performed with the same protocol of sorting out objects in the image having (1) a H coordinate in the range of 5 to 65, a S coordinate in the range of 18 to 82 and a L coordinate in the range of 88 to 145 (for light blue seeds) and a size of more than 700 pixels and objects in the image having (2) a H coordinate in the range of 19-79, a S coordinate in the range of 19-83 and a L coordinate in the range of 62 to 130 (for dark blue seeds) and a size of more than 700 pixels. These two object specifications (1) and (2) were combined to shoot out both, light blue (LB) seeds (1 ) and dark blue (DB) seeds (2) (hence, this is called the LB_DB or DB_LB Shout Out method). The method was performed once (only 1 shoot-out done) using an ASM® EUREKA sorter in order to test the influence of different colored backlight on the purification of the white seeds of a mixed colored seed lot. The percentages (denoted by “P” in the Tables, % impurity) refers to the contamination of blue seeds in the white fraction detected with a multispectral imaging device as a quality control step. The multispectral imaging device in this example was employed with an autofeeder and a blue conveyor belt. The blue color of the conveyor belt as background may hinder the exact distinction of white and light blue seeds. In consequence, some light blue seeds in this analysis may have been classified as white, leading to an overestimation of the purity. Therefore, the purities as specified in Table 1 and Table 2 should be understood as relative purity, with the same systematic bias across all the backgrounds tested, allowing to compare their relative to each other. Control analysis by using the multispectral imaging device may also be performed using different colors for the conveyor belt. A black-colored conveyor belt was found to give the most accurate results with respect to visual inspection. The lower percentages (below P) indicate a higher sorting purity. The amount of “white” seed (denoted by W in the Tables) was measured in grams (g). The non-col- ored “white” seeds, in this example, had a red seed coat. Best backlight may be identified by a high yield with a high purity (i.e. , low percentage of blue seeds) of preferably above 99 %, more preferably above 99.7 %, which can specifically be achieved by repeating performing the method at least 2 or 3 times. The method was performed using approximately 200g of mixed wheat seeds comprising non-colored cereal seeds and seeds containing a blue aleurone. The quality control was done with the multispectral imaging device (videometer®) on approximately 600 seeds.

[0282] Table 1 : Experimental results for Example 1

[0283] Table 1 : Experimental results for Example 1 (continuation)

[0284] These results are summarized in Figures 3A and 3B. As can be seen in the Figures and in above-identified experimental data, colored backlights such as magenta (Hue 300), blue (Hue 240) and cyan (Hue 180) are, in contrast to colored backlights of yellow (e.g. Hue 47) , green (e.g. Hue 71 ) and red (e.g. Hue 340) non-suitable backlight colors to achieve a pure white seed fraction as indicated by a significantly larger percentage of blue seeds remaining in the white fraction after one round of sorting. In general, background colors with Hue values between 150 and 310 are non-suitable colors. Green, yellow and red colored backlight resulted in approx. 2- 5 % remaining impurities of blue seeds in white fraction averaged across the five different seed lots tested (after 1 shoot-out or sorting run, as was used here to test various backlight colors), specifically depending on lightness, whereas the other colors showed 6-25% remaining impurities. Each of the different seed lots used, corresponds to a mixture of blue and non-colored seeds from a different genotype, with all blue seeds containing the same blue aleurone locus.

[0285] Example 2:

[0286] Example 2 essentially corresponds to Example 1 . Example 2 was performed to further test the backlight colors from Experiment 1 with regard to their ability to facilitate a highly accurate sorting, resulting in a relative purity of above 99,7% of the white seed fraction (after at least 2 sortings, compared to other backlight colors), as currently required for certified seed registration. In addition, the best backlight colors from each color group were compared to the previously used backlight colors to evaluate any potential superior performance with regard to purity and / or product yield. Genotypes from 6 different varieties segregating for the blue aleurone color and containing white and blue seeds were sorted with an ASM® EUREKA seed sorter using the same protocol as in Example 1 . The sorting method was repeated at least twice on the white seed fraction and the resulting white fractions were weighed to obtain the product yield and analyzed for its purity using the multispectral imaging device (videometer®). At least 600 seeds were analyzed for each white fraction and the impurity was expressed as percentage of detected blue seeds within the white seeds, and the purity was assessed on approximately 600 seeds of the resulting white fractions using the multispectral imaging device as described in Example 1 .

[0287] Table 2: Experimental results for Example 2

[0288]

[0289]

[0290] Table 2: Experimental results for Example 2 (continuation) Table 3: Experimental results for white and Cyan background and no background lighting (“black") for Example 2

[0291] The results obtained indicate that, for a backlight, in the HSL color space, of a H coordinate of H < 150 or H > 310 (with the exception of H332-338) and a L coordinate of 0.2 < L < 0.80, (see Fig 3a and 3b) the genotypes could be sorted to a purity of 99,7% or higher. Genotype 23NGTA004793 was consistently difficult to sort and the desired purity could only be achieved with a green colored backlight with three consecutive sort outs. When purity was averaged across all tested genotypes, the yellow colored backlight performed best by achieving a relative purity of 99,86% of white seed fractions on average. It should be noted that not all seed lots behave the same. This is probably due to the fact that both, the blue seed fraction and the white seed fraction of some of the different genotypes used, differ significantly in color appearance. For example, genotype 21 NGTA004789 appears significantly darker than e.g. genotype 21 NGTA004785. The different genotypes are derived from different field grown varieties, into which the same blue aleurone locus has been introgressed by re-current backcrossing. Therefore, the different genotypes also differ in size (thousand kernel weight), shape, or degree of shriveling. As a result, the best background light might differ from one genotype to another.

[0292] In addition, depending on the priority of the seed sorting, such as with respect to purity requirement and / or priority of resulting seed quantity (product yield), individual seed lots might be best sorted with a particular background, resulting in the highest product yield and a lower purity of e.g. 99%. One such example could be the yellow color (H,S,L)=(20,255,150) in case of genotype 21 NGTA004791 in which the significantly highest product yield was obtained with two rounds of sorting and a white seed yield exceeding 100 g with a corresponding relative purity of >99%. Consequently, for important individual genotypes, especially of larger volumes, such as several kg or tons, it might be recommended to try a couple of the identified backgrounds on smaller aliquots to identify the optimal background lighting for larger scale sorting. In another example, like in the context of breeding, where hundreds of different small seed lots may need to be sorted, and not all available seeds being needed, a background light, allowing for consistent high purity sorting across all different seed lots, but with a lower product yield might be preferred.

[0293] In comparison to the suitable backlight colors listed in Table 2, applying a white backlight or a cyan backlight, (and cyan is the default color advised for the sorter, and for sorting these types of seeds) did not allow to purify the white fraction of two selected seed lots to sufficient purity. In addition, the amount of white seeds obtained after 2 shots was very low (see Table 3). Due to black painting, the background for the cameras appears black, if the background lights are shut off. This background gave different results, depending on the genotypes sorted. In case of genotype 23NGTA004787 the purity was high and the product yield relatively low, compared to the suitable background colors listed in Table 2. In case of the other genotypes, the purity was low and the product yield very low.

[0294] Example 3:

[0295] Example 3 essentially corresponds to Example 1 . In this example, the method of sorting wheat seed was repeated for a number of 3 shots on different genotypes from which larger amounts of seeds were available. The purity analysis was performed by visual inspection of a total of 4 different images using the multispectral imaging device (videometer ®) taken each time of approx. 170 - 200 seeds from the white fraction which were put into a petri dish and imaged on a black background plate. The amount of blue seeds detected in a total of approximately 700-800 seeds of the white fraction are listed in the table. Cross validation with a genotypic purity assessment, using a SNP marker for the BA locus, confirmed the high level accuracy of this image based purity assessment method. From the available seed lots the two visually lightest and two darkest seed lots were selected for blue-white sorting, in order to cover the entire variation space of seed color.

[0296] In Table 4, the number of white seeds is denoted by #w, the number of blue seeds is denoted by #b, the amount of white seeds (in gram) is denoted by A, the average number of blue seeds is denoted by 0#b and the percentage of impurity is denoted by P.

[0297] Table 4: Experimental results for Example 3

[0298] Example 4: 3n blue aleurone seed depletion using lightness (low L) method

[0299] Principle of the Lightness Method:

[0300] To shoot out the darkest seeds from a seedlot, the Lmin object setting value is put at 0 (maximum darkness) and the Lmax value is chosen according to the percentage of dark blue seeds aimed to be shot out (see the virtual example in Fig. 10). Due to the overlap in darkness of the 3n and 2n Bia fraction (as shown in Fig. 10) the likelihood to shoot out monosomic, 2n (and to a low percentage 1 n) Bia seeds, increases when more dark blue seeds are shot out. With the same (low) Lmax value, more seeds are shot out from a dark seedlot as compared to a light seedlot, hence seedlots with significantly different color expression need different Lmax settings to achieve a shoot out of the same percentage of darkest blue seeds.

[0301] The lightness (or low L) method uses only the lightness aspect of the blue seeds (to be shot- out) to differentially target the darkest blue seeds to shoot out 3n blue aleurone seeds (also named herein double blue (or DB) or disomic seeds (3n refers to the aleurone and disomic / dou- ble refers to the embryo)). To achieve that, the Lmin object value of the protocol is set to 0 and an Lmax value is chosen, which results in the shoot out of the demanded proportion by weight of the darkest seeds. This was tested on seedlots from pooled segregating spikes with an anticipated percentage of double blue seeds of 10 % (pre-sorted to obtain a blue seed fraction using the above methods). HSL setting for the backlight in the ASM® EUREKA color sorter was set at H20 (0-360 scale) / S255 (0-255 scale) / L150 (0-255 scale), with shoot-out object settings : H: 354-54, Smin: 35, Smax: 99, Lmin: 0, Lmax: x (depends on % to be shot out; a lighter seedlot requires a higher Lmax lightness values when compared to a dark seedlot, for the same percentage of shoot out). The results are shown in Table 5 below, showing the Bia locus copy number analysis of blue seeds from segregating spikes of 4 different BLA lines and their corresponding DB shoot out fractions (5%, 10% and 20% by weight), using the Lightness Method.

[0302] Table 5

[0303] In this example, the percentage of double blue seeds in the blue / ”white” seeds from segregating spikes varied between 12% and 27% (17% in average). The efficiency of the Lightness method was approx. 50% at best (about 50 % of 3n blue aleurone in the shoot-out fraction, @5% shoot out by weight of blue seeds).

[0304] The about 20% shoot out by weight is recommended for the blue fraction of seed lots from segregating spikes to deplete for approx. 50% Double Blue seeds with an accompanied loss of approx. 15% of monosomic blue seeds (1 n / 2n blue aleurone). Whereas only 5% to 10% shoot out is recommended on blue seeds from segregating spikes, if the attempt is to maximize main- tainer multiplication rates and efficiencies (less losses of 1n / 2n maintainer seed). It seems that for blue seedlots derived from segregating spides, a 5 - 10 % shoot out (by weight) can contain more than 50 percent of double blue (3n) seeds.

[0305] In further examples using the above lightness method, it was seen that the Lmin at 0 is best used as standard for any new seedlot (works well across different seedlots), the Lmin can also be increased - e.g., the Lmin (object) value can be raised up to 30 for dark seedlots and up to 60 for light seedlots, without significantly influencing the percentage of dark blue seeds being shot out. Also, in this (low L) method, the best Lmax for a 10% dark blue seed shoot is 77 < L < 87, for a 15% shoot out is 79 < L < 95, for a 20% shoot out is 81 < L < 100, for a 30% shoot out is 84 < L < 103, for a 40% shoot out is 87 < L < 107, and for a 50% dark blue seed shoot out is 91 < L < 110 (0-255 L scale).

[0306] Also, when testing other H object settings in the low L method, it was found that essentially any Hue object setting including the range from 354 < H to H <45 also works fine, as well as minor variations of + / - 3 thereof.

[0307] Example 5: 3n blue aleurone shoot-out improvement by use of Lightness and Saturation

[0308] Surprisingly, it turned out that also reducing the Saturation object setting in the Lightness method above improves the visual impression of the dark seed shoot out. The fraction looks often more homogeneous and darker. The mean values of the videometer supported the visual impression. Hence, protocols based on both a low saturation (S) and low lightness (L) object setting are especially suited to shoot out 3n Bia seeds from the blue seed fraction.

[0309] In a first protocol using the backlight HSL settings of H20 / S255 / L150, the Smin and Smin object settings were both set to 0, and depending on the appearance of the seedlot (a lighter seedlot requires higher Smax and Lmax values as compared to a dark seed lot for the same percentage of dark blue shoot out) and the amount of seeds which are intended to be shot out from the blue fraction, the Smax was set between 60 and 78 and the Lmax was set between 78 and 120 (0-255 scale), the H object setting was 354-54. It turned out that the shot-out seeds appeared the darkest, when the Smax value rather than the Lmax value restricts the amount of seeds being shot out with a particular setting and hence is the preferred way to shoot out the desired amount of dark seeds to deplete the maximum of double blue (disomic) seeds.

[0310] When comparing this Lightness and Saturation (or low L+S) method to the Lightness method of Example 4, on the same seedlots of different wheat varieties (containing a mixture of non-col- ored and 1 n; 2n and 3n blue aleurone seeds, 5 varieties produced in Germany and 2 in France), the L+S method gave a darker ejected fraction (as measured with the videometer), indicating a higher proportion of 3n Bia seeds were shot out by the L+S method compared to the Lightness method.

[0311] In a second test, seedlots from pooled segregating spikes were sorted once to split into blue al- uerone and non-colored (lacking blue aleurone, “white”) seeds. The blue fraction was equally split, and approx. 10% of the darkest blue seeds (Shoot out (g)) were shot out either by the Lightness method of Example 4 or the L+S method. The percentage of 3n Bia seeds in the dark blue fraction was determined by genotyping (“% DB” in Table 6 below). From the weights and percentages of 3n Bia seeds in the non-depleted fraction and the dark blue fractions the percent depletion of 3n Bia seeds was calculated.

[0312] In line with the videometer results, for the seedlots in which the amount of 3n Bia seeds was not limited the L + S method works more efficient than the method based on Lightness (low L) only. In such seedlots (here line B and C), within the same / similar amount of seeds shot out (target 10%), a higher proportion of 3n Bia versus 1n and 2n Bia seeds were found with the L+S method as compared to the Lightness method (see Table 6: % DB in shoot out). In consequence, here a higher degree of 3n Bia (DB) seeds depletion was achieved with the L+S method.

[0313] Table 6

[0314] Testing different backlight HSL settings in the L+S method, with a light blue seedlot (23NGTA013748), it was found that the best HSL backlight settings in this method were

[0315] H20 / S255 / L150, H0 / S255 / L110, H111 / S255 / L123, H71 / S255 / L123 and H47 / S255 / L150.

[0316] Also, the low L+S and low L methods were tested on other wheat lines producing blue seeds with 1 n, 2n, and 3n blue aleurone, and “white” seeds without blue aleurone. These seed lots were derived from wheat lines which went through 2 subsequent amplification rounds (without

[0317] DB seed removal) and hence contained a higher percentage of DB seeds. A sample of 135-169 randomly picked seeds from the different 15% or 30% dark blue shoot out fractions was analyzed by PCR for the copy number of the BLA locus. The seeds from the reference aliquot were analyzed to determine the composition of the blue seed fraction prior to the shoot outs. Seeds genotyped as On seeds are “white” seeds containing no BLA locus, seeds genotypes as 1 n or 2n are monosomic seeds (the desired maintainer), seeds genotyped as 3n seeds are disomic (double blue, 3n blue aleurone layer) seeds (undesired).

[0318] Results are shown in the following table 7a.

[0319] Table 7a 5

[0320] For Line ii and Line iii an efficient depletion of DB seeds was achieved, where up to more than 90 % (Line ii) and up to 85% (Line iii) of all shot out seeds at 15 % weight turned out to be 3n Bia seeds. At 30 % shoot out, the DB shoot out efficiency dropped slightly for Line iii to 76 %. These results mean that in case of Line i 45 % (with a % loss of maintainer seeds (1n / 2n) in the shoot-out of 29 %), in case of Line ii 61 % (with a % loss of maintainer seeds (1 n / 2n) in the shoot-out of 4 %) and in case of Line iii 58% (with a % loss of maintainer seeds (1n / 2n) in the shoot-out of 12 %) of all the disomic / 3n blue seeds originally present, could be removed from the blue seed fraction with one shoot out, using either the low L or low L+S method and shooting out 30 % of the darkest seeds by weight. The percentage of DB seed removed by a DB shoot out was calculated as follows: (shoot out weight (g) x percentage DB seed detected in shoot out) I (amount of blue seed used for sorting (g) x percentage DB seed detected therein). Since no strong drop in DB depletion efficiency was observed between 15% and 30% shoot out of dark blue seeds, it seems feasible to further increase the amount of dark blue shoot out more (e.g., 40 % or 50% by weight) to further increase the depletion of disomic (3n) seeds.

[0321] Also, testing other H object settings in the low L+S method, it was found that essentially any Hue object setting including the range from 354 < H to H <45 also works fine, as well as minor variations of + / - 3 thereof.

[0322] The low L+S and low L methods were tested further on two of the same lines as above (Line ii and Line iii in table 7a) and 1 additional other wheat line producing blue seeds with 1 n, 2n, and 3n blue aleurone, and “white” seeds without blue aleurone. All these seedlots were derived from wheat lines which went through 2 subsequent amplification rounds and hence contained a higher percentage of DB seeds. To test further the efficacy of the two methods 40% and 50% of the darkest blue seeds were shot out from Line ii and iii. In case of the other line 20% and 40% of the darkest blue seeds were shout out. A sample between 143-164 randomly picked seeds from the different dark blue shoot out fractions was analyzed by PCR for the copy number of the BLA locus. The seeds from the reference aliquot were analyzed to determine the composition of the blue seed fraction prior to the shoot outs. Seeds genotyped as On seeds are “white” seeds containing no BLA locus, seeds genotyped as 1 n or 2n are monosomic seeds (the desired maintainer), seeds genotyped as 3n seeds are disomic (double blue, 3n blue aleurone layer) seeds (undesired). Results are shown in the following Table 7b. Table 7b

[0323] For Line ii and Line iii, the efficiency of DB depletion increased as compared to the shoot out of 30% for Line ii from 61 % (30% shoot-out) to 72% in case of 50% shoot out with the L+S method and to 66% in case of the Lightness method, and for Line iii from 58% to 86% in case of 50% shoot out with the L+S method and to 71 % in case of the Lightness method. For the new seedlot (Line iv) the DB removal was best at 40% shoot out of the darkest blue seeds both with the L+S and the Lightness method, achieving a depletion of the total amount of DB seeds of 79% and 84%, respectively. Table 7b also shows the % loss (by weight) of maintainer seed (1 n / 2n seed) in the shoot-out, showing that often the L+S method has lower losses of maintainer seed than the lightness method.

[0324] Example 6: light blue (LB) shoot-out method

[0325] The maintainer fraction in a blue aleurone-based 2-line wheat hybrid system should fulfill 2 criteria: a) a low amount of dark blue 3n Bia (disomic) seeds, and b) a low amount of male- sterile “white” seeds. After a first blue aleurone seed shoot out, usually approx. 5-15% of white seeds remain in the blue seed fraction (e.g., using an H71 / S255 / L123 backlight color). A target for the maintainer seed stream is to contain less than 5% “white” male-sterile seeds, hence a sorting step may often be required to remove white seeds. A new concept of a light blue shoot out was tested, in an attempt to not only deplete (further) for 3n Bia (disomic) seeds, but in parallel also reduce the amount of white seeds.

[0326] Based upon the results / knowledge obtained from the low L+S Dark Blue shoot out method, the L and S settings were modified. The attempt was to get a high amount of white seeds in the shoot-out as well as optically only very dark blue seeds in that shoot-out, and no obvious loss of light blue seeds.

[0327] In a first protocol the Smax and Lmax object settings were fixed at 98 and 140, respectively. The Smin and Lmin values were remarkably similar for the extreme seedlots (lightest and darkest) tested. For the darkest seedlot the settings were 55 for Smin and 96 for Lmin whereas for the lightest seedlot the settings were 62 for Smin and 102 for Lmin (H object setting: 345-45 or 354-54). Typically, with one shoot-out of this method the percentage of “white” seeds dropped from 6- 15% to 2-3 % in the blue fraction. The best HSL background settings in this method were : H47 / S255 / L150 for the H:354-54 object setting, and H0 / S255 / L110 for the H:345-45 object setting. In one test, pre-sorted blue seeds (shot-out from a mixture or blue and white, so enriched in blue seeds) from the same wheat lines grown in Gatersleben, Germany and Milly, France, were sorted using different sorting strategies. Sorting strategy A used for the seeds from Milly was the LB-DB shoot-out method of Example 1 , done 4 times, and sorting strategy B for the seeds from Gatersleben was the Light blue shoot-out method, done 2 times, and then the LB-DB method of example 1 once. The results are shown in Table 8 below. Both strategies used the H71 / S255 / L123 backlight settings on the ASM® EUREKA sorter, with the HSL object settings for the LB Shoot Out protocol as follows : H: 348-48, Smin: 55, Smax: 140, Lmin: 48, Lmax: 110 (pixel size 1000), and the HSL object settings for the LB-DB shoot-out method as described in Example 1.

[0328] Table 8

[0329] Hence, the sorting strategy including the light blue shoot out method gave a significantly improved reduction of 3n dark blue seeds (from 38 on average to 19 %).

[0330] In another test, using a pre-sorted rejected blue fraction from an initial sorting to improve white seed purity, it was tested what ranges can be used for the HSL object settings in the LB Shoot out method for different types of blue aleurone seed genotypes, so as to shoot our light blue seeds and retain dark blue and white seeds. The protocol was established as such that the dark blue seeds and the white seeds remain in the kept seed fractions and only the light blue seeds are shot out. Testing was performed as follows: mixed seedlots were sorted into blue and white fraction and the Light Blue shoot out protocol was performed on the (ejected) blue fraction. To test the protocol’s efficiency, the white seeds were removed manually from the retained blue fraction to obtain a true blue videometer value and 2 full petri dishes of this fraction were subjected to videometer analysis to get the mean value per plate. The averaged value of the 2 measurements was compared to the averaged mean videometer value of 2 full petri dishes of the shot out fraction. A significant lower value of the retained blue value versus the value of the shot out fraction indicates a preferential shoot out of the light blue (1 and 2 n blue) seeds. The shoot out with the different Bia lines and the different backlight HSL settings was performed such that the target amount of light blue shoot out (which was estimated from the known amplification rounds performed on these seedlots to be approx. 30%) was adjusted only by either increasing the pixel size (starting with 600) in case of overshooting, or decreasing the pixel size (in case of undershooting) of the object settings. This was tested on 6 seedlots (3 lighter and 3 darker blue aleurone seedlots). The HSL backlight and pixel settings used here that provided a good light blue seed shot-out were H71 / S255 / L123 with pixel size 350-450, such as pixel size 400, H0 / S255 / L110 with pixel size 550 to 650, such as pixel size 600; H20 / S255 / L150 with pixel size 550 to 650, such as pixel size 600, and the H47 / S255 / L123 with pixel size 650 to 750, such as pixel size 700, while white and cyan backlights (with pixel size reduced to 300) did not allow for any good light blue shoot out - even with very low pixel sizes a very low amount of blue seeds was shot out and those seeds contained white seeds, which resulted in very light videometer values (not shown below). See Table 9 indicating the videometer values of a LB Shoot out of 3 light blue (L) or dark blue (D) Bia seedlots (backlight H20 / S255 / L150), and T able 10 showing the testing of different backlight colors for the dark blue D3 line.

[0331] In this LB shoot out method, the Lmin object setting for the light blue seed to be shot out that can be used is: 97<Lmin^110, such as an Lmin of 105, and the Smin object setting can be: 65<Smin^75, such as an Smin of 70, while the Lmax can be: 138-145, such as an Lmax of 140, and the Smax can be: 95-255, such as an Smax of 120.

[0332] It was found that the Hue setting of the object to be sorted-out can be 345<H<45 or 344<H<45, but can be any H range, as long as the range from H>354 and H<45 is included (i.e. , including 354<H<45), e.g. the entire H range (e.g., from 345 to 344).

[0333] Table 9

[0334] Table 10

[0335] **: light values of shot out blue seeds because white seeds occur in the shoot out

[0336] Example 7: sorting disease-infected seeds with dark spots from non-infected wheat seeds

[0337] In this example, 100 g of an ergot infected wheat seed fraction containing ergot fruiting bodies and dark spotted / dark colored seeds (next to non-infected seeds) were mixed with 300 g of a non-infected wheat seed batch from a hybrid. Before mixing, the videometer mean values of 2 full petri dishes of both fractions was determined (ergot infected fraction: videometer value of 0.95, clean seed fraction: videometer value 1 .48). After mixing the two the same was done for the resulting mixed fraction (videometer value mixed fraction: 1.18). The mean value of the ergot infected seed fraction was the lowest, and the mean value of the mixed fraction is in between the (dark) value of the ergot-infected fraction and the (light) value of the clean seed fraction. To test the impact of different backlight color on the effectiveness of removing the ergot fruiting bodies and dark colored seeds from the clean seeds by a dark shoot out protocol (low L / lightness method, HSL object settings: H>354 H<54, Lmin: 0, Lmax: 110, Smin: 50, Smax: 90, pixel size 200 (to cover small dark spots)), one sorting / shoot-out was performed on the ASM® EUREKA sorter, and the shoot out as well as the retained fraction were inspected with the videometer. The obtained values are the averages from the inspection of 2 full plates. An effective shoot out is characterized by a dark shoot out fraction (low videometer value) and light retained fraction. A value of the retained fraction approximately as high as the reference of the clean seeds (1 ,48), indicates an efficient removal of ergot fruiting bodies and dark, colored seeds. Such result was achieved with the H71 (green) and H20 (orange) backlight colors only, and almost with HO (red). For the latter it should be noted, that for a better result the Lmax value could have been slightly increased, to shoot out 77g (as in case of H71 green and with an increase of Lmax to 117 for H20 orange) and not only 61g.

[0338] However, the lightness of the retained fraction resulting from the shootout with either cyan or white backlight are significantly lower (1 ,33 and 1 ,38) than the reference and the fractions were also visually less clean (still containing some fruiting body pieces and colored dark seeds), despite the fact that the shootout amount was higher. This result (see Table 11 ) clearly demonstrates the superior performance of the red, green or orange backlights for the removal of ergot and / or dark colored infected seeds versus the standard cyan or white backlights.

[0339] Table 11

[0340] Example 7: Sorting-out darker red rice from a red rice sample

[0341] The preferred backlight HSL settings with the low L object settings as used for sorting out dark blue aleurone seeds were also tested to get a more uniform rice seed batch from a commercial package of 500 grams red rice grains (having lighter and darker red seed color). The lightness (low L) method was applied to shoot out a proportion (about 12%) of the darkest seeds (containing more colorants / pigments such as anthocyanins and / or proanthocyanidins) from the red rice batch to give it more homogeneous appearance. The protocol settings were derived from pic- tures of red rice grains, taken in the ASM® EUREKA sorter and adapting those in a way, that all grains in the picture were recognized by the settings. The protocol was then modified by altering the Lmin to 0 and the Lmax to a value leading to the shoot out of approximately 60 gram of the darkest red seeds. To compare the efficiency of various different backlight colors / HSL settings, the videometer value of the shot-out fraction, as well as the videometer value of the resulting / re- tained fraction was determined (values are given in averaged mean values of 2 petri dishes (approx. 620 seeds each)). The lower the videometer value of the shoot-out with the same weight, or the higher the value of the resulting / retained fraction respectively, the more effective the chosen backlight is for this purpose. The Table 12 below shows the results.

[0342] The backlights with H71 , H 111 , HO, or H20 all resulted in improved sorting of dark red rice removal, resulting in an overall more homogenous and slightly lighter appearing red rice sample, as compared to the cyan or white backlights.

[0343] The backlights as indicated by H value only in the above Table were the ones described in Figure 4-9 so H71 / S255 / L123 or H20 / S255 / L150 (cyan backlight was H180 / S255 / L128 and white backlight was H0 / S255 / L255),

[0344] List of reference numbers sorting device cereal seed non-colored cereal seed seed containing a blue aleurone seed feeder seed stream sorting station feed hopper chute vibratory feeder backlight device front light device camera first backlight device first camera first side second side second backlight device second camera ejector pneumatic ejector target chute sort-out chute controller supplying a seed stream taking an image identifying seeds to be sorted out ejecting seeds identified to be sorted out angle around the center distance from the center box x axis y axis mixed fraction of cereal seeds first backlight ejected fraction of seeds retained fraction of seeds second backlight first round of sorting second round of sorting waste fraction of cereal seeds third round of sorting fourth round of sorting fourth backlight third backlight

Claims

Claims1 . A method of sorting seeds, wherein the seeds contain darker colored and lighter colored seeds or non-colored seeds and colored seeds, the method comprising: i. supplying a seed stream (120) to a sorting station (122), the sorting station (122) comprising at least one backlight device (130) for backlighting a seed of the seed stream (120) and at least one camera (132) for taking at least one image of the backlighted seed; ii. taking, with the camera (132), at least one image of the backlighted seed of the seed stream (120); iii. automatically identifying, from the image taken in step ii., seeds to be sorted out from the seed stream (120); and iv. automatically ejecting seeds identified to be sorted out from the seed stream (120), wherein the at least one backlight, in the HSL color space, has a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80.

2. The method according to the preceding claim, wherein the seeds to be sorted comprise, specifically are, cereal seeds (112), specifically lighter colored and / or non-colored cereal seeds and colored cereal seeds, more specifically lighter colored and / or non-colored cereal seeds and darker-colored cereal seeds.

3. The method according to any one of the preceding claims, wherein the seeds contain lighter colored seeds and darker colored seeds, or non-colored seeds and colored seeds.

4. The method according to any one of the preceding claims, wherein, in step iii., seeds to be sorted out from the seed stream (120) are colored or dark colored seeds identified by identifying, in the image, objects cumulatively fulfilling the following conditions: the objects have predefined color coordinates, specifically color coordinates in a predefined range in the HSL color space, and the objects have one or more of a predefined area, a predefined size, a predefined diameter, a predefined equivalent diameter and a predefined shape.

5. The method according to any one of the preceding claims, wherein the at least one backlight, in the HSL color space, has a H coordinate in at least one range selected from the group consisting of:0 < H < 25 or H > 310, specifically excluding a range of 332 < H < 338; 70 < H < 150;25 < H < 70.

6. The method according to any one of the preceding claims, wherein the at least one backlight, in the HSL color space, has a S coordinate of 0.25 < S < 1 .0, specifically of 0.5 < S < 1 .0, more specifically of 0.75 < S 1 .0.

7. The method according to any one of the preceding claims, wherein the at least one backlight, in the HSL color space, has at least one color selected from the group consisting of: a S coordinate of 0.25 < S < 1.0 and a H coordinate of 25 < H < 70, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.35 < L < 0.78; a S coordinate of 0.33 < S < 1.0 and a H coordinate of 70 < H < 150, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.28 < L < 0.63; a S coordinate of 0.40 < S < 1.0 and a H coordinate of 25 < H or H > 310, wherein the at least one backlight, in the HSL color space, has a L coordinate of 0.13 < L < 0.78.

8. The method according to anyone of the preceding claims, wherein the method is a continuous method, wherein, in step i., a continuous seed stream is supplied to the sorting station (122), wherein, in step ii., a continuous stream of images is taken of the seed stream (120), and, wherein, in step ill., the stream of images is continuously evaluated for continuously identifying seed to be sorted out from the seed stream (120).

9. The method according to any one of the preceding claims, wherein a batch of seeds is provided, and wherein the batch of seeds is subjected to method steps i.-iv. repeatedly, wherein, in each repetition, the batch is diminished by the seeds ejected in step iv. of the previous run.

10. The method according to any one of the preceding claims, wherein the method further comprises at least one second spectral seed sorting step, wherein the second spectral seed sorting step comprises determining at least one item of spectroscopic information on the seeds of the seed stream (120), such as near-infrared spectroscopic information, wherein the at least one item of spectroscopic information is used for automatically identifying seeds to be sorted out from the seed stream (120).11 . The method according to any one of the preceding claims, wherein the method comprises performing steps i. to iv. with at least one first backlight (202), the first backlight (202), in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the method further comprises repeating steps i. to iv. using one of an ejected fraction of seeds (204) or a retained fraction of seeds (206) with at least one second backlight (208), the second backlight (208), in the HSL color space, having a H coordinate of H < 150 or H > 310 and a L coordinate of 0.10 < L < 0.80, wherein the second backlight (208) is different from the first backlight (202).

12. The method according to the preceding claim, wherein the method comprises separately repeating steps i. to iv. with the ejected fraction of seeds (204) and the retained fraction of seeds (206).

13. The method according to any one of the preceding claims, for sorting out colored seeds from a seed mixture containing darker and lighter colored seeds and non-colored seeds,wherein a sorting protocol uses a combination of a) the HSL object settings for the darker colored seeds as provided by the sorting device (110) and b) the HSL object settings for the lighter colored seed as provided by the sorting device (110), which method can be performed once or can be repeated several times, such as repeating 1-3 times, to maximize non-colored seed purity.

14. The method according to any one of claims 1 to 11 for sorting out darker colored seeds from a seed mixture containing darker and lighter colored seeds, which mixture may contain non-colored seed, wherein the L object settings as provided by the sorting device (110) for the darker colored seeds, are modified by lowering the Lmin object settings to at least the value of the darkest seed to ensure a shoot-out of the darkest seeds, and wherein the Lmax is set to a value resulting in the shoot-out of a desired % by weight of the darkest seeds, and wherein the H object setting is any H range comprising 354<H<45, such as wherein the Lmin is 0 or 0<Lmin^60 or 0<Lmin^30, and such a method wherein the backlight is as described in any one of the preceding claims, or is a white LED backlight or a backlight with an L coordinate of L = 255 (white).

15. The method of any one of the preceding claims, wherein the colored seed or darker colored seed is a disease-infected seed with darker seed color or darker spots compared to the non-infected seed that is lighter or has no darker spots, such as ergot-infected cereal seeds.

16. A sorting device (110) for sorting seeds, comprising:I. at least one seed feeder (118) for supplying a seed stream (120) to at least one sorting station (122); andIL the at least one sorting station (122), comprising at least one backlight device (130) for backlighting seeds of the seed stream (120), the sorting station (122) further comprising at least one camera (132) for taking at least one image of the backlighted seed, and the sorting station (122) further comprising at least one ejector (146) for ejecting seeds from the seed stream (120), wherein the sorting device (110) is configured for performing the method according to any one of the preceding claims.

17. A computer program comprising instructions which, when the program is executed by the sorting device (110) according to any one of the preceding claims referring to a sorting device (110), causes the sorting device (110) to perform the method according to any one of the preceding claims referring to a method.

18. A computer-readable storage medium, comprising instructions which, when the instructions are executed by the sorting device (110) according to any one of the preceding claims referring to a sorting device (110), cause the sorting device (110) to perform the method according to any one of the preceding claims referring to a method.

19. A use of the sorting device (110) according to any one of the preceding claims referring to a sorting device (110) for a purpose of use, selected from the group consisting of: sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds; sorting out of dark colored seeds from a mixture of seeds containing dark colored seeds and lighter colored seeds (with or without non-colored seeds in said mixture); sorting out of dark colored cereal seeds from a mixture of seeds containing dark colored cereal seeds and lighter colored cereal seeds (with or without non-colored cereal seeds in said mixture); sorting out of colored cereal seeds from a mixture of cereal seeds (112) containing non-colored cereal seeds (114) and colored cereal seeds; sorting out disease- infected seeds with darker seed color from non-infected seed; sorting out of colored cereal seeds containing a blue aleurone (116) from a mixture of cereal seeds (112) containing non-colored cereal seeds (114) and cereal seeds containing a blue aleurone (116); sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 70 < H < 150; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 25 < H < 70; sorting out of colored seeds from a mixture of seeds containing non-colored seeds and colored seeds using at least one backlight having, in the HSL color space, a H coordinate of 0 < H < 25 or H > 310.

Images