Grain seed sorting

Abstract

A method of sorting cereal seeds (112) is disclosed. The cereal seeds (112) comprise colourless cereal seeds (114) and seeds (116) containing a blue aleurone layer, or comprise coloured seeds containing a dark blue aleurone layer and seeds containing a light blue aleurone layer. The method comprises: i. supplying a seed stream (120) to a sorting station (122) comprising at least one backlight device (130) for back-illumination of the seeds (112) of the seed stream (120) and at least one camera (132) for taking at least one image of the back-illuminated seeds; ii. taking at least one image of the back-illuminated seeds of the seed stream (120) using the camera (132); iii. automatically identifying from the image taken in step ii. seeds to be sorted out of the seed stream (120); and iv. automatically ejecting the identified seeds to be sorted out of the seed stream (120), wherein the at least one backlight has an H coordinate of H < 150 or H > 310 and an L coordinate of 0.10 < L < 0.80 in the HSL colour space.

Description

Technical Field

[0001] This invention relates to a method for sorting grain seeds, a sorting apparatus for sorting grain seeds, and the uses of the sorting apparatus. Specifically, the method and the sorting apparatus can be used to sort grain seeds comprising a mixture of (darker) colored grain seeds and (lighter) or colorless grain seeds, such as grain seeds with a colorless aleurone layer and seeds containing a blue aleurone layer in the field of plant breeding. However, other applications, including grain seed sorting, are also possible. Background Technology

[0002] In plant breeding, such as cereal plant breeding (e.g., wheat breeding), blue aleurone layer loci and / or genes can be useful color markers to indicate other genes and / or loci of interest located on the same chromosome, or even on the same chromosome arm, as indicated by the color marker. Therefore, the presence of blue in the seed aleurone layer can indicate the presence of the gene of interest. Generally, the closer the two genes and / or loci are, the more reliable the blue indication of the gene of interest, as greater distances typically increase the incidence of recombination and / or breakage. As an example, a seed might contain a recessive male sterility gene and / or locus that causes male sterility in plants lacking a restorer gene, and the gene of interest is a male fertility restorer gene linked to the blue aleurone layer color locus. Blue can be used to distinguish fertile “blue” seeds containing a blue aleurone layer locus indicating the presence of a restorer gene from colorless “white” sterile seeds lacking a blue aleurone layer locus indicating the absence of a restorer gene. Therefore, seeds containing male-sterile loci and / or genes, and fertility-restoring genes linked to the blue aleurone layer locus, can be used as maintainer lines to propagate both male-sterile "white" seeds and fertile "blue" seeds, as their self-pollinated offspring contain both types of seeds. These sterile seeds can be grown into useful maternal lines for hybrid seed production because they only produce fruit after cross-pollination of (male) fertile plants. As another example, the blue aleurone layer locus can be used in cereal seeds (such as wheat seeds) to identify higher anthocyanin content in the seeds and / or to create new foods with natural coloring, making synthetic pigment products (e.g., in breakfast cereals) redundant. Therefore, in any production field or seed batch containing both seeds with the blue aleurone layer and "white" seeds, it may be necessary to separate the colored "blue" seeds from the colorless "white" seeds.

[0003] WO 2023 / 088892 A1 describes a method for classifying / sorting seeds, comprising the steps of: providing a sample containing at least one seed; obtaining near-infrared (NIR) spectra of at least a subset of the sample; determining, based on the obtained NIR spectra, the presence of an organic colorant in at least that subset of the sample; and classifying / sorting, based on that determination, at least that subset of the sample. Based on this, white seeds with incorrect coloration and blue seeds with faded blue coloration can be classified. Furthermore, it is said that single-blue and double-blue seeds can also be distinguished.

[0004] WO 2014 / 109993 A2 describes a system and method for separating seeds or kernels based on optical differences in starch composition. The method for separating seeds or kernels based on optical differences in starch composition includes receiving a seed group containing a plurality of seeds. The method further includes illuminating each seed in the seed group from an illumination source positioned behind the seeds, such that the seeds are backlit. The method further includes sorting each seed in the seed group based on differences in starch composition. In some cases, the method includes sorting each seed by separating the seed group into groups of glutinous seeds and non-glutinous seeds.

[0005] Despite the advantages of known methods and equipment, several technical challenges remain. Standard separation machines that separate colored seeds based on size and / or density may often be ineffective, as colored seeds are the same size and / or density as normal seeds. Therefore, color separation may be necessary. However, color separators used in existing seed processing equipment to separate colored seeds from normal seeds typically provide lower sorting quality. Color separators use an electronic eye to identify different colors. Seeds pass through the electronic eye, and if a seed with a different color from the desired seed is identified, a sudden burst of air can be activated to push that seed into a rejection box, while the remaining seeds can go into another box. Therefore, there remains a need to improve the sorting efficiency of optical separators, particularly in terms of sorting purity and / or sorting yield. In particular, there remains a need to sort seeds that carry only one dose (1n) of the BLA locus, which often exhibit only weak coloration, frequently resulting in inadequate seed sorting.

[0006] The problem to be solved

[0007] Therefore, it is desirable to provide apparatus and methods that at least partially address the aforementioned technical challenges of known devices and methods. Specifically, one object of the present invention is to provide apparatus and methods for improving the sorting purity and sorting yield when sorting grain seeds comprising a mixture of grain seeds having normal (colorless) or light-colored seeds and (darker) colored seeds, such as grain seeds having a normal (colorless) aleurone layer and grain seeds containing a blue aleurone layer, or diseased grain seeds having a different seed color compared to uninfected grain seeds (e.g., darker seed color, slightly reddish seed color, brown / black / gray spots, etc.), or sorting darker seeds (e.g., containing more anthocyanins) from lighter-colored seeds (to obtain grain seeds with more uniform color).

[0008] Furthermore, in the aforementioned hybridization system of cereals, the blue aleurone layer color can be used to distinguish fertile "blue" seeds containing a blue aleurone layer locus (indicating the presence of a restorer gene) from colorless "white" sterile seeds lacking a blue aleurone layer locus (indicating the absence of a restorer gene). It should be noted that the aleurone layer is triploid (3n). Due to the possible paternal transmission of the blue aleurone layer locus, seeds may contain one (1n), two (2n), or three (3n) copies of the blue aleurone layer (BLA or BA) locus in the aleurone layer, which will result in light blue (1n or 2n) seeds and dark blue (3n) seeds, respectively. In addition, repeated breeding of blue seed batches containing 3n blue aleurone layer seeds (also known as disomy or double blue seeds) results in a significant reduction in the proportion of sterile ("white") and maintainer line (1n or 2n blue) seeds (because 3n BLA seeds only produce (3n) blue seed offspring), and thus leads to a decrease in the yield of white seeds and maintainer line seeds per unit area. On average, across different varieties, after three generations of breeding, the proportion of white seeds in harvested seed batches decreases from approximately 64% to 25%. To maximize the production efficiency of useful seeds for hybridization and preservation, the amount of dark blue 3n seeds should be reduced through a specific seed sorting step, in which either the darkest seeds in the blue seed fraction are blown out, or alternatively, only the light blue seeds are blown out and only the light blue fraction is used for seed propagation. This sorting step can also be used to remove (darker) colored seeds from seed batches containing both (darker) colored and colorless or light-colored seeds. Summary of the Invention

[0009] This problem is solved by a method for sorting grain seeds with the features of the independent claims and a sorting device for sorting grain seeds. Advantageous embodiments that can be implemented independently or in any arbitrary combination are set forth in the dependent claims and throughout the specification.

[0010] A first aspect of the invention is a method for sorting grain seeds, wherein the grain seeds comprise light-colored grain seeds and / or colorless grain seeds, as will be further defined in detail below, as well as colored seeds, particularly those containing a blue aleurone layer, or the grain seeds may have been pre-sorted to remove colored seeds from colorless seeds, such that the grain seeds comprise at least light blue (1n or 2n blue aleurone layer) colored seeds and dark blue (3n aleurone layer) colored seeds, and may also comprise colorless seeds.

[0011] As used herein, the term "sorting" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or customary meaning. The term may specifically refer to, but is not limited to, the process of differentiation and separation. Sorting may include at least two process steps, which may be performed in parallel and / or in a time-overlapping manner, wherein, in a first step, objects may be differentiated from each other according to one or more sorting criteria, and wherein, in a second step, objects may be separated from each other based on that differentiation. Specifically, the differentiation of objects may include assigning objects to at least two different sorting categories. Separation may be performed based on sorting categories, wherein objects assigned to a first sorting category may be physically separated from objects assigned to a second sorting category. The grain seed sorting described herein can be performed on any seed sorting equipment that allows for adjustable background light color, such as the ASM® EUREKA sorter, the Bühler® Sortex H sorter, the H-series color sorters from AnySort (Anhui Jiexun Optoelectronic Technology Co., Ltd.), and / or the 3U Vision® OPTICA sorter.

[0012] The sorting can specifically include sensor-based sorting. Specifically, the differentiation can include using one or more sensor devices to sense sorting criteria. The sorting can include optical sorting. Thus, as an example, the differentiation of objects to be sorted can be performed at least partially by optical means. The sorting can include differentiating objects to be sorted based on color and / or at least one color coordinate, as will be further detailed below, and therefore can include determining at least one color and / or at least one color coordinate of these objects and assigning the objects to at least two different sorting categories based on the determined color and / or color coordinate. The sorting can specifically include automatic sorting of objects to be sorted, particularly without requiring manual operation and / or user interaction, such as by using a combination of one or more sensor devices for achieving differentiation and one or more controllers for achieving separation (such as by actuating at least one separation device). The result of the sorting can include at least two different groups of objects, wherein each object in a particular group may share at least one common attribute or characteristic with other objects in that particular group. Specifically, the sorting can result in the objects to be sorted being separated into at least two distinct groups, wherein objects in the first group may have a specific color and / or at least one specific color coordinate, and objects in the second group may be different from that specific color and / or that at least one specific color coordinate. For example, the sorting may include distinguishing colorless grain seeds as defined herein from colored grain seeds (such as grain seeds containing a blue aleurone layer), separating colorless grain seeds from colored grain seeds (such as grain seeds containing a blue aleurone layer), or separating dark blue seeds from light blue seeds (and any remaining colorless seeds), or separating light blue seeds from dark blue seeds (and any remaining colorless seeds).

[0013] As used herein, the term "seed" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a particular or customary meaning. Specifically, the term may refer to, but is not limited to, the reproductive unit of a flowering plant capable of developing into another such plant by producing functional pollen or male gametes. Thus, the term may specifically refer to fertile seeds. The term may also refer to sterile seeds. The term "sterile seed" in this invention may refer to a seed from which the plant grows is unable or partially unable to produce functional pollen or male gametes (also known as male sterility). This can be attributed to a naturally or artificially introduced genetic predisposition or artificial intervention in the plant in the field. Male sterility / fertility in cereals (e.g., wheat) can be reflected by the seed set rate during self-pollination, for example, by bagging the ears to induce self-fertilization. Similarly, fertility restoration can also be described by comparing the seed set rate of a male-sterile plant crossed with a plant carrying a functional restoration gene to the seed set rate produced by crossing (or self-pollinating) a fully fertile plant. The paternal parent (or pollen parent) is the parent plant that provides male gametes (pollen) for fertilization, while the maternal parent (or seed parent) is the plant that provides female gametes for fertilization; the maternal parent is the plant that produces (hybrid) seeds. Male sterility can be restored, for example, by introducing a functional restoration gene into the genome of a sterile plant.

[0014] As used herein, the term "grain seed" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or customary meaning. The term may specifically refer to, but is not limited to, the seeds of grains. Specifically, grain seeds may include the grains of grains. As used herein, the term "grain" may be equivalent to seed and may include grains used for sowing and / or planting crops. Grain seeds may be whole grains. Grain seeds may specifically be grains comprising the endosperm, germ, aleurone layer, seed coat, and pericarp. The grain seed may be a grain selected from the group consisting of: maize; rice; wheat; barley; sorghum; millet; oats; rye; triticale. The grain seed may specifically be a wheat grain. As used herein, “wheat” can refer to plants from the genus *Triticum*, including but not limited to common / bread wheat (*Triticum aestivum* or *T. aestivum*), emmer wheat (*T. dicoccum*), midge wheat (*T. monococcum*), durum wheat (*T. durum*), kalam wheat (*T. turanicum*), or spelt wheat (*T. spelta*), specifically the hexaploid *T. aestivum* or *T. spelta*, and the tetraploid *T. durum*, including wheat referred to as hard wheat or soft wheat (based on endosperm texture), winter wheat or spring wheat (based on sowing season), and red wheat or white wheat (based on seed coat color). The terms “cereal seed” and “seed” are used interchangeably herein.

[0015] Grain seeds can include light-colored and dark-colored grain seeds, or colorless and colored grain seeds.

[0016] As used herein, the term "colorless" can specifically refer to, but is not limited to, the natural phenotype of cereal seeds that do not possess a blue aleurone layer. Colorless seeds can specifically be seeds with a natural seed coat color and normal aleurone and pericarp colors, such as wheat seeds with white or red seed coats, lacking a blue aleurone layer, and without a purple pericarp (e.g., examples of wheat varieties with white seed color are Blini (spring wheat) and Herroldo (winter wheat), and examples of wheat varieties with red / slightly red seed color are Granary (spring wheat) and Solehio (winter wheat). As used herein, natural or normal cereal seed / pericarp / aleurone layer color refers to the seed / pericarp / aleurone layer color of most cereal crops commercially grown for the production of food, feed, or beverages, excluding niche / niche market cereal seeds with purple or blue seed colors. Colorless seeds may also be referred to herein as “white” seeds, specifically based on seed coat color, and also as known red or white wheat seeds (such as hard red winter wheat or soft white spring wheat) to distinguish them from seeds with a blue aleurone layer or a purple pericarp. Colorless wheat seeds may specifically include red seeds. Colorless seeds may specifically be seeds obtained as in standard cereals (such as wheat), seeds obtained through breeding and / or infiltration from plants lacking a blue aleurone layer, or seeds lacking a blue aleurone layer as used in current commercial wheat in current farming practices or major markets. Colorless seeds may include normal wheat seeds, which are generally referred to as white or red seeds. The normal color of commercially available wheat seed types may be white seed categories and / or red wheat seed categories, specifically excluding purple or blue wheat seeds.

[0017] As used herein, the term "colored" can specifically refer to, but is not limited to, seeds having a deeper color than colorless or normally colored seeds, such as seeds with a deeper seed color than colorless seeds or seeds with darker spots. For example, colored seeds can include seeds with a blue aleurone layer and no purple pericarp color. Alternatively or additionally, colored seeds can include seeds infected with diseases, such as seeds with fungal, bacterial, or viral infections that alter seed color to a darker shade (visible as darker (dark brown, black, or gray) spots on the seed, seeds that are generally darker, or seeds whose color changes from white to red / pink, such as those infected with clover (such as C. purpurea), smut, black spot, black smut, etc., including any secondary infections). In one embodiment, diseased cereal seeds can refer to cereal seeds with dark brown or black spots on the seed. As used herein, whether a seed is colored or darker compared to a (colorless or lighter-colored) control seed can be determined by visual observation or by measurement using a video measuring instrument, colorimeter, or spectrophotometer (e.g., by measuring brightness). In one embodiment of the invention, the sorting method described herein can be used as a quality control tool to quantify the amount of disease-infected (such as ergot-infected) or darker-colored cereal seeds in a seed batch. This can determine whether the seed batch meets the requirements of a specific market / channel, such as whether it is acceptable for release as a food or feed product, or whether it needs to be destroyed or redirected to other (e.g., non-food / non-feed) markets / channels. Specifically, colored seeds can be seeds with a natural seed coat color and a blue aleurone layer and normal pericarp color, such as wheat seeds with a white or red seed coat color, a blue aleurone layer, and no purple pericarp. The blue aleurone layer color can be associated with the presence of anthocyanins in the cereal seed aleurone layer. The blue color of the seed may be genetically controlled by the blue aleurone layer locus involved in anthocyanin biosynthesis. Current standard commercial wheat seeds may not contain a blue aleurone layer, but this trait can be introduced into wheat through introgression from some species of the Triticeae family, or by adding blue aleurone loci to plants lacking these loci through plant transformation or (targeted) genome editing. Several blue aleurone loci that confer the blue aleurone trait may have been transferred into wheat from Triticeae species such as *Thinopyrum ponticum*, *Agropyron elongatum*, *Triticum boeoticum*, *Triticum monococcum*, or *Thinopyrum bessarabicum*. In this paper, seeds containing a blue aleurone layer may also be referred to as “blue” seeds (or BLA seeds).As will be further elaborated below, colored seeds can specifically refer to both light-colored seeds, such as those with 1n or 2n BLA loci in the aleurone layer, and dark-colored seeds, such as those with 3n BLA loci in the aleurone layer. In one embodiment, colored cereal seeds as used herein may also have a purple pericarp and a normal (non-blue) aleurone layer, and colorless cereal seeds as used herein may have a natural seed color, such as white or red seed color (with a normal (non-purple) pericarp and (non-blue) aleurone layer color). Colored cereal seeds may also have a darker seed color, such as dark red rice seeds, while colorless cereal seeds may have a lighter seed color, such as light red rice seeds, wherein darker seeds can be separated from lighter seeds to obtain a more uniform seed batch. Alternatively, colored seeds can be a subfraction of seeds containing residues of undesirable portions of the seeds during processing, such as aleurone layer residues in polished seeds (e.g., rice) or seed coat residues in peeled seeds (e.g., peanuts).

[0018] As used herein, the term "blue aleurone layer locus" (or BLA locus) is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or customary meaning. Specifically, the term may refer to, but is not limited to, the genetic locus that causes the blue aleurone layer phenotype, which can be transferred to wheat through introgression from related species. Seeds containing a blue aleurone layer are seeds containing the BLA locus, specifically wheat seeds containing the BLA locus, and may be referred to as blue seeds.

[0019] For example, cereal seeds can include wheat seeds. A 2-line male-sterile system can be used with a maintainer line plant that produces white sterile and blue fertile cereal seeds upon self-pollination. Any conventional wheat plant can be used as a paternal parent to restore fertility. The BLA locus and restorer gene can be located on a monosomy (42+1 Chr plants) or on a partially homologous chromosome pair (42 Chr plants). The BLA locus and restorer gene can be located on the same arm or different arms of the same chromosome. The aleurone layer can be triploid, such as being part of the endosperm of a triploid (3n), and due to the possible paternal transmission of the BLA locus, the seed may contain 1 (1n), 2 (2n), or 3 (3n) copies of the BLA locus in the aleurone layer, which would result in light blue, blue, or dark blue seeds, respectively. For possible embodiments of this two-line male-sterile system, see, for example, Whitford et al., 2013, J. Exp. Botany [Journal of Experimental Botany] 64 (18): 5411–5428, and Zhou et al., 2006, CropScience [Agro-Crop Science] 46:250-255, CN100420368, 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 monosomy, or on the same arm of one of the two chromosomes in a partially homologous chromosome pair. Thus, these genes may be closely linked, or alternatively, the BLA locus and the restorer gene may each be located on a different chromosome arm of the same monosomy or partially homologous chromosome. As an example, the restorer gene may be MS1, MS5, MS9, MS22, MS26, or MS45, depending on the cause of male sterility. For instance, if a mutation, inactivation, or deletion of the MS1 gene causes male sterility, then MS1 ​​may be the restorer gene to be used, and if a mutation, inactivation, or deletion of all MS45 genes (on each wheat subgenome (A, B, and D)) causes male sterility, then MS45 may be the restorer gene to be used. Alternatively or additionally, the blue aleurone layer gene / locus may be one of the genes / loci described in US11,390,877 B2, WO 2019 / 043082 A1, or WO 2020 / 056259 A1.The blue aleurone layer locus can be obtained from, for example, tall icegrass, fluffy icegrass, wild emmer wheat, emmer wheat, sand emmer wheat, common wheat or long spike wheatgrass, or wheat lines with the BLA locus of gene introgression, or from the known seed accession numbers Sebesta Blue, Blue Sando, Blue Baart, Blue Onas, Blue 1, PBB or Blue Norco.

[0020] In one embodiment of the invention, the male-sterile maternal plant of the invention may comprise triple homozygous mutations of the MS45 male-fertile polynucleotide in wheat, which cause a male-sterile phenotype. In one embodiment, such a male-sterile plant may be obtained from a two-line hybrid system in wheat comprising triple homozygous mutations of the MS45 male-fertile polynucleotide (such as EMS mutations that inactivate each MS45-A, MS45-B, and MS45-D gene, or mutations that inactivate each MS45-A, MS45-B, and MS45-D gene obtained by genome editing methods), and a plant restoration donor chromosome component comprising a 4E chromosome component from the genera *Thinopyrum* or *Agropyron*, the 4E chromosome component comprising: (a) plant polynucleotides conferring plant seed phenotypes (such as seed color, e.g., blue aleurone layer, P gene, anthocyanins, or Kala4); and (b) The MS45 male fertility restoration locus restores male fertility in ms45 male-sterile wheat plants via the 4E chromosome component, wherein the expression of the 4E chromosome component is functionally complementary to the male-sterile phenotype derived from the triple homozygous MS45 mutation, making the wheat plant male-fertile. In one embodiment, the plant polynucleotide conferring plant phenotypes (such as seed color) may be located on the same chromosome arm (not separated by the centromere) as the MS45 male fertility restoration locus on the 4E chromosome component. In one embodiment, the MS45 hybridization system and the male-sterile maternal wheat plant may be as described in WO 2020056259.

[0021] The method includes the following steps, which can be performed in a given order. However, different orders are also possible. In particular, one, more than one, or all of the method steps can be performed once or repeatedly. Further, these method steps can be performed sequentially, or alternatively, one or more method steps can be performed in a time-overlapping manner or even in a parallel and / or combined manner. The method may further include additional method steps not listed.

[0022] The method includes:

[0023] i. Supplying a seed stream to a sorting station, the sorting station including at least one backlighting device for backlighting the seeds of the seed stream and at least one camera for capturing at least one image of the backlit seeds.

[0024] ii. Use the camera to take at least one image of the backlit seeds of the seed stream;

[0025] iii. Automatically identify the seeds to be sorted from the seed stream from the images captured in step ii; and

[0026] iv. Automatically eject the identified seeds to be sorted from the seed stream.

[0027] The at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space.

[0028] Specifically, the at least one backlight may have an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.20 ≤ L ≤ 0.80 in the HSL color space.

[0029] As used herein, the term "supply" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, processes that enable further processing. Supply may specifically include providing a seed stream to a sorting station.

[0030] As used herein, the term "seed stream" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, multiple seeds. A seed stream may include multiple seeds, wherein these multiple seeds may be provided to a sorting station in a directed manner. Additionally or alternatively, a seed stream may include multiple seeds arranged in a regular manner. A seed stream may include multiple seeds, wherein the seeds of a seed stream may be arranged individually in a row or a line. Thus, as an example, the seeds of a seed stream supplied to a sorting station may include multiple seeds provided to the sorting station one by one.

[0031] As used herein, the term "sorting station" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a device or combination of devices configured to perform at least one sorting function. Specifically, a sorting station may include at least one sensor device, such as a camera, configured to distinguish seeds in a seed stream. A sorting station may further include at least one ejector configured to separate seeds in a seed stream by ejecting seeds to be sorted from the seed stream. The sensor device and ejector of the sorting station may be connected to each other, for example via one or more controllers, such that when the sensor device distinguishes seeds in the seed stream, the ejector ejects the corresponding seeds from the seed stream. The sorting station may further include at least one front illumination device for illuminating the seed stream from the front. A sorting station may include two, four, or even more front illumination devices. The front illumination devices may include white LEDs.

[0032] As used herein, the term "backlight" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or customary meaning. The term may specifically refer to, but is not limited to, light provided from the rear. Backlight may specifically be or include light present in the background. As used herein, the term "light" may refer to, but is not limited to, electromagnetic radiation in the visible light spectrum. In this document, the term "visible spectral range" generally refers to the spectral range of 380 nm to 760 nm. The method may include the use of different backlights, such as using different backlights in different (specifically repeated) sorting steps. Where the method may include more than one backlight, the backlights involved in the same sorting step may have a constant light setting.

[0033] As used herein, the term "backlight device" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a device configured to generate light in the sense defined above. A backlight device may specifically include at least one light source for generating light. For example, a backlight device may include at least one light-emitting diode (LED). A backlight illumination device may specifically include multiple LEDs, such as at least one red LED, at least one green LED, and at least one blue LED. These multiple LEDs may form a colored white LED configured to emit light at least in the visible spectrum. The colored white LED may be configured to provide different colors in the visible spectrum based on color mixing from the multiple LEDs. Color mixing in the colored white LED can be controlled by individually controlling the multiple LEDs. The backlight device may be arranged behind a seed stream in the camera's field of view and thus provide "backlighting".

[0034] Colored backlighting can be provided directly by a backlighting device, specifically by controlling multiple LEDs of different colors to achieve the corresponding backlight color, such as tunable RGB LEDs. Alternatively or additionally, colored backlighting can be provided using a combination of monochromatic or white light and one or more color filters to achieve the corresponding backlight color. Therefore, in this example, the backlighting device may additionally include one or more color filters. Alternatively or additionally, a colored background can be used, which is illuminated by a monochromatic or white backlighting device. For example, a colored background panel, sheet, or conveyor belt can be arranged behind the seed stream to provide the corresponding backlight color.

[0035] As used herein, the term "camera" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, a device having at least one imaging element configured to record or capture spatially resolved one-dimensional, two-dimensional, or even three-dimensional optical data or information. As an example, a camera may include at least one camera chip, such as at least one CCD chip and / or at least one CMOS chip configured to record images. As used herein, without limitation, the term "image" may specifically refer to data recorded using a camera, such as multiple electronic readings from an imaging device, such as pixels of a camera chip.

[0036] In addition to at least one camera chip or imaging chip, a camera may also include other elements, such as one or more optical elements, for example, one or more lenses. As an example, the camera may be a fixed-focus camera having at least one lens that is fixedly adjusted relative to the camera. Alternatively, however, the camera may also include one or more variable lenses that can be adjusted automatically or manually.

[0037] Specifically, the camera can be a color camera. Therefore, for each pixel, color information can be provided or generated, such as color coordinates for three colors, e.g., H (hue), S (saturation), L (luminance) and / or R, G, B. A greater number of color values ​​are also possible, e.g., four colors per pixel, e.g., R, G, G, B. Color cameras are well known to those skilled in the art. Thus, as an example, each pixel of the camera chip can have three or more different color sensors, such as color recording pixels, e.g., one pixel for red (R), one pixel for green (G), and one pixel for blue (B). For each pixel, e.g., for R, G, B, values ​​can be recorded by the pixel, e.g., numerical values ​​in the range of 0 to 255, depending on the intensity of the corresponding color. As an example, in addition to using color triplets such as H, S, L and / or R, G, B, quadruplets such as R, G, G, B or C, M, Y, K, etc., can be used. The color sensitivity of a pixel can be generated by a color filter or by the appropriate inherent sensitivity of the sensor elements used in the camera pixel. These techniques are well known to those skilled in the art.

[0038] As used herein, the term "capturing at least one image" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, one or more of imaging, image recording, image acquisition, and image capture. The term "capturing at least one image" can include capturing a single image and / or multiple images, such as an image sequence. For example, capturing an image may include continuously recording a sequence of images, such as video or film. Capturing at least one image can be initiated automatically, for example, upon automatic detection of a seed stream within and / or a predetermined area of ​​the camera's field of view. These automatic image acquisition techniques are known, for example, in the field of automatic barcode readers, such as from automatic barcode reading applications. As an example, capturing an image can be performed by acquiring an image stream or "live stream" with a camera, wherein one or more images are (specifically automatically) stored and used as at least one image of a backlit seed. Image acquisition may be supported by a controller, such as by at least one processor of the controller, and the storage of images for image processing and / or evaluation may be performed on the controller's data storage device.

[0039] As described above, from the image captured in step ii., the seeds to be sorted are automatically identified, and the identified seeds to be sorted from the seed stream are automatically ejected. As used herein, the term "automatic" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or customary meaning. Specifically, the term may refer to, but is not limited to, processes performed entirely by means of at least one computing unit, particularly those requiring no manual operation and / or user interaction. The term "automatic" may specifically refer to any process performed by means of a controller.

[0040] As used herein, the term "identification" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, the process of identifying one or more features in an image. Identification may specifically include at least one image processing step for determining seeds to be sorted in an image. For example, identification may include at least one image processing step for identifying seeds in the image. Identification may further include at least one image processing step for determining whether a seed identified in the image is a seed to be sorted. Optionally, the identification may further include at least one object detection step, specifically at least one moving object detection step, such as for estimating the motion of seeds in a seed stream, for example, for estimating the motion of seeds to be sorted in the seed stream.

[0041] As used herein, the term "seed to be sorted" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, seeds in a seed stream that are to be separated from other seeds in the seed stream. Seeds to be sorted can be predefined as being separated from other seeds in the seed stream, for example, by defining a specific phenotype of the seed that will be separated from other seeds in the seed stream that have a different phenotype. For example, seeds to be sorted from the seed stream may specifically include (darker) colored seeds, such as seeds containing a blue aleurone layer.

[0042] As used herein, the term "ejection" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, the process of removing objects from a set of objects. Specifically, ejection may include removing seeds to be sorted from a seed stream. Ejection may include removing seeds to be sorted by means of at least one of mechanical and pneumatic devices. For example, a pneumatic ejector may be configured to eject seeds to be sorted from a seed stream using compressed air, such as separating seeds to be sorted from a seed stream by using compressed air directed via a nozzle. By ejecting the seeds to be sorted from the seed stream, the seeds in the seed stream can be separated into a first set of seeds including the seeds to be sorted and a second set of seeds including the seeds not ejected from the seed stream.

[0043] As described above, the at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. As used herein, the term "color space" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, any coordinate system by which the color of an object (such as the color of a seed or the color of a backlight) can be characterized, such as mathematically or physically. Various color coordinate systems are commonly known to those skilled in the art, such as the color coordinate system defined by the CIE (International Commission on Illumination). Other color coordinate systems besides those defined by the CIE are also feasible. Color coordinates can span or define a color space in their entirety, such as by defining three or four basis vectors. Therefore, when a camera captures an image of an object, the camera generates a value for each color coordinate for each pixel. As an example, the camera chip may contain a color sensor for recording the value of each color, such as HSL and / or RGB (red, green, blue) and / or L... a Triples like b, or quadruples like CMYK (cyan, magenta, yellow, black), where these values ​​depend on the sensitivity of the camera chip.

[0044] As used herein, the term "color coordinates" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, coordinates used in any color coordinate system for describing color using coordinates. Several color coordinate systems are commonly known to those skilled in the art and may also be used in the context of this invention. Thus, by way of example, chromaticity coordinate systems or human-perception-based coordinate systems, such as the CIE 1964 color space, the Munsell color system, or other coordinate systems, such as H, S, L and / or R, G, B and / or L, a, b, may be used.

[0045] As used herein, the term "HSL color space" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, the cylindrical coordinate representation of a point in the RGB color model. Specifically, the HSL color space can be defined by color coordinates hue (H), saturation (S), and lightness (L). As used herein, the term "hue" may refer to, but is not limited to, color coordinates describing the angle around the central vertical axis of the color cylinder of the HSL color space. Hue can specifically be an angle from 0 to 360 degrees, where H=0=360 defines red, where H=30 defines orange, where H=60 defines yellow, where H=90 defines yellow-green, where H=120 defines green, where H=150 defines cyan, where H=180 defines cyan, where H=210 defines cyan-blue, where H=240 defines blue, where H=270 defines cyan-magenta, where H=300 defines magenta, and where H=330 defines magenta-red. As used herein, the term "saturation" can refer to, but is not limited to, a color coordinate describing the distance from the central vertical axis of the color cylinder in the HSL color space. Saturation can be defined using relative terms, specifically from 0 to 1, and / or absolute terms on a scale from 0 to 255, where a relative term of 1 for full saturation corresponds to an absolute value of 255. Saturation describes the ratio of a color's colorfulness to its lightness. Full saturation can be described by a value of 1 or alternatively by a value of 255, and zero saturation can be described by a value of 0. Saturation can include percentages from 0% to 100%, equivalent to a scale range from 0 to 1, indicating a balance between pure color and white. For example, for a hue of pure green (H=120), 100% saturation can consist of only green light, specifically without red or blue light. Therefore, changing the saturation of a primary color within a percentage range is equivalent to increasing or decreasing the saturation of the other two primary colors by the same amount. A pure green hue (H=120) with 50% saturation can indicate that red and blue are each increased by 50%. As used herein, the term "luminance" can refer to, but is not limited to, a color coordinate describing the distance along the central vertical axis of the color cylinder in the HSL color space. Luminance can be defined by relative terms, specifically from 0 to 1, and / or by absolute terms on a scale from 0 to 255, where a full luminance of 1 corresponds to an absolute value of 255. Luminance can describe the lightness or darkness of a color relative to the brightness of white under similar lighting. A full luminance value of 1 can correspond to white, where a luminance value of 0 can correspond to black. Unless otherwise indicated, HSL values ​​in this document can be described on a hue scale of 0 ≤ H < 360, a relative luminance scale of 0 ≤ L ≤ 1, and a relative saturation scale of 0 ≤ S ≤ 1.Therefore, the S and L values ​​given in the range of 0 to 1 or as % are always on a relative scale, while other values, specifically integer values ​​greater than 1, are always on an absolute scale of 0 to 255 for S and L, or on an absolute scale of 0 to 260 for H.

[0046] Without narrowing the scope of the invention, this invention will be described specifically with respect to the HSL color space. However, it should be noted that other color spaces (such as those mentioned above and others) are also feasible. For example, colors in the RGB color space (where R, G, B ∈ [0, 1]) can be converted to colors in the HSL color space according to the following equation:

[0047] (1),

[0048] If H < 0 in equation (1), then

[0049] (2),

[0050] (3),

[0051] in and In equations (2) and (3), S and L ∈ [0, 1].

[0052] Similarly, colors in the HSL color space (where H ∈ [0, 360) and S, L ∈ [0, 1]) can be converted to colors in the RGB color space according to the following equation:

[0053] (4),

[0054] (5),

[0055] (6),

[0056] (7),

[0057] (8),

[0058] (9).

[0059] Therefore, any color or color coordinate given in the HSL color space below can be equivalently described by a color or color coordinate in the RGB color space and / or any other color space, and vice versa. For example, color converters on the World Wide Web can also be used to convert colors from one color space to another, such as convertacolor.com, or, for example, the color converter for RGB to HSL conversion located at www.w3schools.com / colors / colors_hsl.asp, the color converter for RGB to HSL conversion located at www.rapidtables.com / convert / color / rgb-to-hsl.html, and / or the color converter for HSL to RGB conversion located at www.rapidtables.com / convert / color / hsl-to-rgb.html.

[0060] The at least one backlight may specifically have an H coordinate in the HSL color space that is within at least one range selected from a group of the following:

[0061] -0 ≤ H < 25 or H ≥ 310, specifically excluding the range 332 ≤ H ≤ 338;

[0062] -70 ≤ H ≤ 150;

[0063] -25 ≤ H ≤ 70.

[0064] Specifically, H coordinates within the range of 0 ≤ H < 25 or H ≥ 310 may include red backlighting. H coordinates within the range of 70 ≤ H ≤ 150 may include green backlighting. H coordinates within the range of 25 ≤ H ≤ 70 may include yellow backlighting.

[0065] The at least one backlight may have an S-coordinate of 0.25 ≤ S ≤ 1.0, specifically 0.5 ≤ S ≤ 1.0, and more specifically 0.75 ≤ S 1.0 in the HSL color space.

[0066] The at least one backlight may have an H coordinate in the range of 0 ≤ H ≤ 20, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.78 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 20 ≤ H ≤ 25, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.36 ≤ L ≤ 0.78 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 25 ≤ H ≤ 45, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.63 ≤ L ≤ 0.78 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 45 ≤ H ≤ 50, an S coordinate in the range of 0.65 ≤ S ≤ 1.0, and an L coordinate in the range of 0.56 ≤ L ≤ 0.72 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 50 ≤ H ≤ 55, an S coordinate in the range of 0.6 ≤ S ≤ 1.0, and an L coordinate in the range of 0.44 ≤ L ≤ 0.72 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 55 ≤ H ≤ 60, an S coordinate in the range of 0.6 ≤ S ≤ 1.0, and an L coordinate in the range of 0.35 ≤ L ≤ 0.56 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 60 ≤ H ≤ 70, an S coordinate in the range of 0.6 ≤ S ≤ 1.0, and an L coordinate in the range of 0.35 ≤ L ≤ 0.63 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 70 ≤ H ≤ 85, an S coordinate in the range of 0.6 ≤ S ≤ 1.0, and an L coordinate in the range of 0.44 ≤ L ≤ 0.50 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 85 ≤ H ≤ 100, an S coordinate in the range of 0.6 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.63 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 100 ≤ H ≤ 105, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.55 in the HSL color space.Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 105 ≤ H ≤ 125, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.50 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 125 ≤ H ≤ 150, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.50 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 310 ≤ H ≤ 332, an S coordinate in the range of 0.5 ≤ S ≤ 1.0, and an L coordinate in the range of 0.13 ≤ L ≤ 0.45 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 338 ≤ H ≤ 350, an S coordinate in the range of 0.8 ≤ S ≤ 1.0, and an L coordinate in the range of 0.35 ≤ L ≤ 0.50 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 338 ≤ H ≤ 350, an S coordinate in the range of 0.3 ≤ S ≤ 1.0, and an L coordinate in the range of 0.55 ≤ L ≤ 0.78 in the HSL color space. Alternatively or additionally, the at least one backlight may have an H coordinate in the range of 350 ≤ H ≤ 360, an S coordinate in the range of 0.5 ≤ S ≤ 1.0, and an L coordinate in the range of 0.28 ≤ L ≤ 0.78 in the HSL color space.

[0067] The at least one backlight can have an S-coordinate of 0.25 ≤ S ≤ 1 and an H-coordinate of 25 ≤ H ≤ 70 in the HSL color space. Therefore, for a yellow backlight, the saturation of the backlight can be in the range of 0.25 ≤ S ≤ 1. Further, the at least one backlight can have an L-coordinate of 0.35 ≤ L ≤ 0.78 in the HSL color space. The at least one backlight can have an H-coordinate within 45 ≤ H ≤ 50, an S-coordinate within 0.65 ≤ S ≤ 1, and an L-coordinate within 0.55 ≤ L ≤ 0.65 in the HSL color space. For example, the at least one backlight can have an H-coordinate of 47, an S-coordinate of 1.0, and an L-coordinate of 0.59 in the HSL color space. This particular yellow can correspond to the RGB color (R, G, B) = (255 / 210 / 46). Compared to other backlight colors, these specific yellows achieve optimal purity and lower product yield for light-colored seed batches, and very good purity and good product yield for dark-colored seed batches, as will be further detailed below.

[0068] Additionally or alternatively, the at least one backlight may have an S-coordinate of 0.33 ≤ S ≤ 1.0 and an H-coordinate of 70 ≤ H ≤ 150 in the HSL color space. Therefore, for a green backlight, the backlight saturation can be in the range of 0.33 ≤ S ≤ 1.0. Further, the at least one backlight may have an L-coordinate of 0.28 ≤ L ≤ 0.63 in the HSL color space. The at least one backlight may have an H-coordinate within 109 ≤ H ≤ 113, an S-coordinate within 0.3 ≤ S ≤ 1.0, and an L-coordinate within 0.45 ≤ L ≤ 0.50 in the HSL color space. For example, the at least one backlight may have an H-coordinate of 111, an S-coordinate of 1.0, and an L-coordinate of 0.48 in the HSL color space. This particular green may correspond to the RGB color (R, G, B) = (37, 245, 0). Alternatively or additionally, the at least one backlight may have an H coordinate of 101, an S coordinate of 1.0, and an L coordinate of 0.45 in the HSL color space. This particular green may correspond to an RGB color of (R, G, B) = (73, 230, 0). Compared to other backlight colors, these particular green backlights can achieve very good purity and lower product yield, as will be further detailed below. Alternatively or additionally, the at least one backlight may have an H coordinate of 69 ≤ H ≤ 73, an S coordinate of 0.6 ≤ S ≤ 1.0, and an L coordinate of 0.45 ≤ L ≤ 0.50 in the HSL color space. For example, the at least one backlight may have an H coordinate of 71, an S coordinate of 1.0, and an L coordinate of 0.48 in the HSL color space. This particular green may correspond to an RGB color of (R, G, B) = (78, 245, 0). Compared to other backlight colors, these specific green backlights achieve optimal overall purity and lower product yield, as will be further detailed below.

[0069] Additionally or alternatively, the at least one backlight may have an S-coordinate of 0.40 ≤ S ≤ 1.0 and an H-coordinate of 25 ≤ H or H ≥ 310 in the HSL color space. Therefore, for a red backlight, the backlight saturation can be in the range of 0.40 ≤ S ≤ 1.0. Further, the at least one backlight may have an L-coordinate of 0.13 ≤ L ≤ 0.78 in the HSL color space. The at least one backlight may have an H-coordinate within 338 ≤ H ≤ 342, an S-coordinate within 0.3 ≤ S ≤ 1.0, and an L-coordinate within 0.65 ≤ L ≤ 0.70 in the HSL color space. For example, the at least one backlight may have an H-coordinate of 340, an S-coordinate of 0.75, and an L-coordinate of 0.68 in the HSL color space. This particular red backlight can correspond to an RGB color of (R, G, B) = (234 / 108 / 150). Alternatively or additionally, the at least one backlight can have an H coordinate of 340, an S coordinate of 0.78, and an L coordinate of 0.68 in the HSL color space. This particular red can correspond to an RGB color of (R, G, B) = (255, 87, 143). Alternatively or additionally, the at least one backlight can have an H coordinate of 340, an S coordinate of 0.71, and an L coordinate of 0.68 in the HSL color space. This particular red can correspond to an RGB color of (R, G, B) = (231, 111, 151). Alternatively or additionally, the at least one backlight can have an H coordinate of 340, an S coordinate of 0.71, and an L coordinate of 0.68 in the HSL color space. This particular red can correspond to the RGB color (R, G, B) = (231, 111, 151). Alternatively or additionally, the at least one backlight can have an H coordinate of 0, an S coordinate of 1.0, and an L coordinate of 0.43 in the HSL color space. This particular red can correspond to the RGB color (R, G, B) = (219, 0, 0). These particular red backlights can achieve good purity and high product yield, as will be further detailed below.

[0070] In one embodiment of the invention, for the low L or low L+S method described herein (for blowing out darker seeds, such as dark blue seeds), a backlight in the HSL color space described herein can be used, but a white LED backlight or a backlight with an L coordinate of L = 255 (white) can also be used. Therefore, whenever the backlight in the low L / brightness method or low L+S method described herein is mentioned, such a white LED backlight or a backlight with an L coordinate of L = 255 (white) can be used as an alternative backlight.

[0071] As described above, step iii. includes automatically identifying seeds to be sorted from the seed stream from the image captured in step ii. In step iii., the seeds to be sorted from the seed stream can be identified by recognizing objects in the image that cumulatively satisfy the following conditions:

[0072] These objects have predefined color coordinates, specifically color coordinates within a predefined range in the HSL color space, and

[0073] These objects have one or more of the following: predefined area, predefined size, predefined diameter, predefined equivalent diameter, and predefined shape.

[0074] Specifically, seed identification can include determining the color coordinates and area of ​​objects in an image, where

[0075] - If the determined color coordinates are within a predefined range in the HSL color space, then the object is determined to have predefined color coordinates; and

[0076] - If the determined area exceeds the area threshold, the object is determined to have a predefined area.

[0077] In the HSL color space, this predefined range can have H coordinates from 0 to 85, specifically from 5 to 65 and / or from 19 to 79, S coordinates from 15 to 85, specifically from 18 to 82 and / or from 19 to 83, and L coordinates from 60 to 150, specifically from 91 to 141 and / or from 66 to 130. The area threshold can be in the range of 200 to 1500 pixels, such as from 400 to 1000 or from 500 to 1000 pixels, specifically in the range of 600 to 850 pixels, more specifically in the range of 650 to 700 pixels, and most specifically 700 pixels (generally, for smaller seeds, the pixel size is best reduced to ensure coverage of the relevant area). For example, when the corresponding object can have a size greater than 700 pixels, an object in the image with an H coordinate in the range of 19 to 79, an S coordinate in the range of 19 to 83, and an L coordinate in the range of 66 to 130 can be identified as a seed to be sorted from the seed stream. Alternatively or additionally, when the corresponding object can have a size greater than 700 pixels, an object in the image with an H coordinate in the range of 5 to 65, an S coordinate in the range of 18 to 82, and an L coordinate in the range of 91 to 141 can be identified as a seed to be sorted from the seed stream. The HSL coordinates can give L or S values ​​on an absolute scale from 0 to 255 (which can also be converted to a relative scale from 0 to 1, as will be apparent to those skilled in the art), and H values ​​on an absolute scale from 0 to 360.

[0078] Alternatively or additionally, the identification of seeds to be sorted may include determining at least one identification parameter, which includes weighting color coordinates by the area of ​​an object in the image, specifically by using the product of the color coordinates and the area of ​​the object in the image. Thus, if the identification parameter exceeds a certain threshold, an object in the image can be identified as a seed to be sorted from the seed stream.

[0079] In one embodiment, the method of the present invention may include determining object settings (as provided by a sorting device) for the darkest available seed batch and determining object settings (as provided by a sorting device) for the lightest available seed batch. These two object settings are combined as two different rules into a single sorting procedure and tested against “white seeds” in the seed batch. The L min setting of the light blue rule may be increased and / or the pixel size of both rules may be adjusted to ensure that white seeds are not (or very limitedly) blown out from any blue / white seed mixture and that all blue seeds are effectively blown out. For example, the method can be performed on a batch of seeds containing dark blue and light blue aleurone layer seeds and colorless "white" seeds, using the same procedure to sort out objects in the image that have (1) H coordinates in the range of 5 to 65, S coordinates in the range of 18 to 82, and L coordinates in the range of 88 to 145, and a size greater than 700 pixels (for sorting light blue seeds), and objects in the image that have (2) H coordinates in the range of 19 to 79, S coordinates in the range of 19 to 83, and L coordinates in the range of 62 to 130, and a size greater 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 following HSL backlight settings: H340 / S192 / L172, H71 / S255 / L123, or H0 / S255 / L110. In one embodiment of the method, the image has H coordinates in (1) and (2) that include 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 blow out light blue (LB) seeds and dark blue (DB) seeds (this is referred to as the LB_DB or DB_LB blowing method). The method can be performed once or can be repeated multiple times, such as 1-3 times (therefore, a total of 2-4 sortings), for example, to maximize the purity of colorless seeds.

[0080] This method can specifically be a continuous method. As used herein, the term "continuous" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a particular or customary meaning. The term can specifically refer to, but is not limited to, a process attribute that is uninterrupted in time. Specifically, a continuous method for sorting grain seeds can include performing these method steps repeatedly and at least partially overlapping in time. In step i., a continuous stream of seeds can be supplied to a sorting station. In step ii., a continuous stream of images of the seed stream can be captured. In step iii., the image stream can be continuously evaluated to continuously identify seeds to be sorted from the seed stream.

[0081] In this method, a batch of seeds can be provided. As used herein, the term "batch" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or customary meaning. Specifically, the term can refer to, but is not limited to, a collection of objects produced jointly, or a collection of objects produced and mixed at different times and / or locations. Specifically, this batch of seeds can include multiple seeds produced jointly (specifically, with regard to at least one of the production location and production time). This batch of seeds can be repeatedly subjected to method steps i.-iv., specifically at least twice. In each repetition, the batch may be reduced due to the seeds sprayed in the previous round of step iv. Therefore, by repeatedly performing method steps i.-iv. on this batch of seeds, the purity of the seeds passing through the sorting station can be improved.

[0082] The method of the present invention may further include at least one backlight identification step, specifically at least one backlight identification step prior to step i. This backlight identification step may include identifying at least one backlight used for sorting. Thus, in the backlight identification step, at least one backlight used for sorting in steps i. to iv. can be determined. This backlight identification step may include multiple sorting operations using multiple backlights (specifically, different backlights). Each sorting operation may include using an aliquot of the seeds to be sorted. At least one backlight may be determined via at least one quality control step. This quality control step may include evaluating sorting purity across multiple backlights and optionally further considering the product yield of the sorted seeds by weighing the corresponding aliquot fractions. Alternatively, the quality control step may include evaluating sorting purity and the amount of loss of desired seeds during sorting across multiple backlights. This method can be advantageously used to further improve sorting efficiency in large-scale seed production by using specific backlights best suited for particular seed production (within the HSL / RGB range provided herein).

[0083] Specifically, the method may include performing steps i. to iv. using at least one first backlight, the first backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The method may further include repeating steps i. to iv. on either the ejected seed fraction or the retained seed fraction using the first backlight or using at least one second backlight, the second backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The second backlight may be different from the first backlight. Thus, for example, the method may include at least two consecutive sorting rounds, wherein a first sorting round may include using the first backlight, and a second sorting round may include using the second backlight, the second sorting round being performed using the ejected seed fraction or the retained seed fraction. Furthermore, a third or even more sorting rounds may be performed on either the ejected seed class or the retained seed class. The third sorting round may include using the same first backlight or another backlight, which will be further detailed below.

[0084] The first backlight can be used for the primary purpose of sorting grain seeds. The primary purpose of sorting grain seeds may include at least one purpose selected from the group consisting of: maximizing the yield of colorless grain seeds; maximizing the yield of (darker) colored grain seeds; maximizing the purity of colorless or lighter colored grain seeds, specifically by removing impurities such as (darker) colored seeds, weed seeds, diseased seeds, fungal sclerotia (such as ergot fruiting bodies) and / or other plant parts formed on the ear; maximizing the purity of (darker) colored grain seeds; reducing the number of (darker) colored seeds among colorless or lighter colored seeds; reducing the number of lighter or colorless seeds among (darker) colored seeds; maximizing the purity of colorless seeds while minimizing colorless seed loss; maximizing the purity of colored seeds while minimizing colored seed loss; or any combination thereof.

[0085] The second backlight can be used for a secondary purpose of sorting grain seeds. The secondary purpose of sorting grain seeds may include at least one purpose selected from the group consisting of: maximizing the yield of colorless or lighter-colored grain seeds; maximizing the yield of (darker) colored grain seeds; maximizing the purity of colorless or lighter-colored grain seeds; maximizing the purity of (darker) colored grain seeds; reducing the number of (darker) colored seeds among colorless or lighter-colored seeds; reducing the number of colorless or lighter-colored seeds among (darker) colored seeds; maximizing the purity of colorless seeds while minimizing colorless seed loss; maximizing the purity of colored seeds while minimizing colored seed loss; or any combination thereof.

[0086] The first backlight can be used for the primary purpose of sorting grain seeds with a blue aleurone layer or sorting seeds lacking a blue aleurone layer. The primary purpose of sorting grain seeds can include at least one purpose selected from the group consisting of: maximizing the yield of colorless grain seeds lacking a blue aleurone layer; maximizing the yield of colored grain seeds with a blue aleurone layer; maximizing the purity of colorless grain seeds lacking a blue aleurone layer, specifically by removing impurities such as colored seeds with a blue aleurone layer, weed seeds, diseased seeds, and / or other plant parts; maximizing the purity of colored grain seeds with a blue aleurone layer; reducing the number of colored seeds with a blue aleurone layer among colorless seeds lacking a blue aleurone layer; reducing the number of colorless seeds lacking a blue aleurone layer among colored seeds with a blue aleurone layer; maximizing the purity of colorless seeds while minimizing the loss of colorless seeds; maximizing the purity of blue aleurone layer seeds while minimizing the loss of blue aleurone layer seeds; or any combination thereof.

[0087] The second backlight can be used for the secondary purpose of sorting grain seeds with a blue aleurone layer or sorting seeds lacking a blue aleurone layer. The secondary purpose of sorting grain seeds can include at least one purpose selected from the group consisting of: maximizing the yield of colorless grain seeds lacking a blue aleurone layer; maximizing the yield of colored grain seeds with a blue aleurone layer; maximizing the purity of colorless grain seeds lacking a blue aleurone layer, specifically by removing impurities such as colored seeds with a blue aleurone layer, weed seeds, diseased seeds, and / or other plant parts; maximizing the purity of colored grain seeds with a blue aleurone layer; reducing the number of colored seeds with a blue aleurone layer among colorless seeds lacking a blue aleurone layer; reducing the number of colorless seeds lacking a blue aleurone layer among colored seeds with a blue aleurone layer; maximizing the purity of colorless seeds while minimizing colorless seed loss, and maximizing the purity of colored seeds while minimizing colored seed loss; or any combination thereof.

[0088] Specifically, the second purpose of sorting grain seeds can be different from the first purpose of sorting grain seeds.

[0089] Repeating steps i through iv using either the ejected seed fraction or the retained seed fraction may include using a sorting procedure different from the initial sorting in step iii. This sorting procedure may define parameters based on which seeds to be sorted are identified in the image. Thus, as an example, in step iii., seeds to be sorted from the seed stream can be identified by recognizing objects in the image with color coordinates of a predefined sorting procedure, specifically objects with color coordinates within a predefined range in the HSL color space. This can be used to sort dark and light seeds from colorless seeds, such as sorting (dark and light) blue aleurone layer seeds from colorless seeds, or to sort dark seeds from light or colorless seeds, such as sorting 3n dark blue aleurone layer seeds from light blue aleurone layer seeds or colorless seeds, or to sort light seeds from dark and colorless seeds, such as sorting light blue aleurone layer seeds from dark blue aleurone layer and colorless seeds. In the following text, the HSL object settings (for the seeds to be blown out) are provided in absolute scale, but as mentioned above, they can also be equivalently converted to relative scale. Predefined sorting procedures may include at least one procedure with HSL object settings for the seeds to be ejected / blown out, these settings being selected from the group consisting of: H coordinates including a range of 354 ≤ H ≤ 45 or H ≥ 354 and H ≤ 54, S coordinates in S... min ≤ S ≤ S max Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The H coordinate includes the range of 354 ≤ H ≤ 45 or the range of H ≥ 345 and H ≤ 79, and the S coordinate is within the range of S... min ≤ S ≤ S max Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The H coordinate includes the range of 354 ≤ H ≤ 45 or the range of H ≥ 345 and H ≤ 45, and the S coordinate is within the range of S... min Within the range of ≤ S ≤ 98, where S min In 55 ≤ S min Within the range of ≤ 62, and the L coordinate is within L min Within the range of ≤ L ≤ 140, where L min In 96 ≤ L min The range of H coordinates is ≤ 102; the range of H coordinates includes 354 ≤ H ≤ 45 or H ≥ 354 and H ≤ 54; the range of S coordinates is within S... min Within the range of ≤ S ≤ 98, where S min In 55 ≤ S min Within the range of ≤ 62, and the L coordinate is within L min Within the range of ≤ L ≤ 140, where L min In 96 ≤ L min Within the range of ≤ 102; the combination of the first color coordinates and the second color coordinates, wherein the first color coordinates include H coordinates within the range of 354 ≤ H ≤ 45 or 19 ≤ H ≤ 79, S coordinates within the range of 19 ≤ S ≤ 83, and L coordinates within the range of 62 ≤ L ≤ 130, and the second color coordinates include H coordinates within the range of 354 ≤ H ≤ 45 or 5 ≤ H ≤ 65, S coordinates within the range of 18 ≤ S ≤ 82, and L coordinates within the range of 88 ≤ L ≤ 145; H coordinates within the range of 354 ≤ H ≤ 45, S coordinates within the range of 35 ≤ S ≤ 99, and L coordinates within the range of L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30 and where L max It is set to achieve a certain percentage of blowout, such as 78 ≤ L max L ≤ 127 max The H coordinate includes the range of 354 ≤ H ≤ 45, and the S coordinate is within the range of S... min ≤ S ≤ Smax Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to achieve a certain percentage of blowout, such as 78 ≤ L max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The H coordinate includes the range 354 ≤ H ≤ 45, and the L coordinate is within the range L. min ≤ L ≤ L max Within the range, where L min 96 ≤ L min ≤ 110, 96 ≤ Lmin ≤ 102 or 97 ≤ Lmin ≤ 110, such as L of 105 min And L max 138 ≤ Lmax ≤ 145, such as L of 140 max And the S coordinate is in S min ≤ S ≤ S max Within the range, where S min 55 ≤ S min ≤ 75, 55 ≤ S min ≤62 or 65 ≤ S min ≤ 75, such as 70 for S min And S max 95 ≤ S max ≤ 255, such as 98 or 120 S max Alternatively, the predefined sorting procedure may include at least one procedure with HSL object settings for seeds to be ejected / blown, wherein the H object setting can be any H range as long as it includes 354 ≤ H ≤ 45, such as a sorting procedure with H object settings for any H range for seeds to be ejected / blown but including the range 354 ≤ H ≤ 45, S min and S max The seeds are supplied by a sorting device for the seeds to be blown out, and L min =0 or 0 ≤ L min ≤ 30 and L max Between 85 and 127.

[0090] Additionally, seeds to be sorted can be further identified by recognizing objects in the image that have one or more of a predefined area, predefined size, predefined diameter, predefined equivalent diameter, and predefined shape. For example, seeds to be sorted can be further identified by recognizing objects in the image whose determined area exceeds an area threshold, which is 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 pixels, and most specifically in the range of 200, 400, 600, 700, or 750 pixels. Compared to seed batches with an average thousand-seed weight (TKW) of around 30 (300-600 pixels), larger seeds with an average TKW above 40 typically have a higher pixel range (700-1500 pixels), while smaller pixel sizes (such as 200) are used to blow out seeds with black spots. Furthermore, screening or separating seeds based on size / weight prior to the sorting method of this invention can help improve sorting accuracy and reduce losses.

[0091] Specifically, this method may include repeating steps i. to iv. separately for the ejected fraction and the retained fraction of the seed. For example, steps i. to iv. may be repeated (once or multiple times, such as 1-3 times) for the retained fraction using a first backlight, while steps i. to iv. may be repeated for the ejected fraction using a second backlight. The retained fraction may consist mostly of colorless seeds, such as sterile “white” seeds lacking a blue aleurone layer, while the ejected fraction may consist mostly of colored seeds, such as fertile seeds with a blue aleurone layer. In the following, the first and second backlights, as well as further optional backlights, may be given in absolute HSL values ​​(360 scale for H, and 255 scale for S and L). However, these color values ​​may be equivalently converted to relative HSL values, or even converted to different color spaces, such as the RGB color space as described above. In the above text, the first backlight may have at least one color in the HSL color space selected from the group consisting of: H coordinate 340, S coordinate 192, and L coordinate 172; H coordinate 71, S coordinate 255, and L coordinate 123. The second backlight may have at least one color in the HSL color space selected from the group consisting of: H coordinate 20, S coordinate 255, and L coordinate 150; H coordinate 111, S coordinate 255, and L coordinate 123; H coordinate 0, S coordinate 255, and L coordinate 110; or any H coordinate differing from the above H coordinates by + / - 3 (within the 0-360 scale), any S coordinate differing from the above S coordinates by + / - 15 (within the 0-255 scale), or any L coordinate differing from the above L coordinates by + / - 5 (within the 0-255 scale).

[0092] Alternatively or additionally, steps i. to iv. can be performed using a first backlight, which may have a color in the HSL color space, for example, an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110 (e.g., to maximize the yield of colorless (such as sterile "white") seeds). Steps i. to iv. can be repeated for the seed retention fraction using a second backlight different from the first backlight (and which may have at least one color in the HSL color space, for example, an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172, or an H coordinate of 71, an S coordinate of 255, and an L coordinate of 123). Steps i. to iv. can be repeated for the seed ejection fraction using a third backlight, which may have an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The third backlight may be different from both the first and second backlights. In this example, the first backlight may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110 in the HSL color space, or an H coordinate of 71, an S coordinate of 255, and an L coordinate of 123. The second backlight may have at least one color in the HSL color space selected from the group consisting of: an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172; or an H coordinate of 71, an S coordinate of 255, and an L coordinate of 123. Steps i to iv may be repeated at least once (e.g., 1-3 times) on the retained fraction of the seed using any second backlight. The third backlight can have at least one color in the HSL color space selected from the group consisting of: 20 H coordinate, 255 S coordinate and 150 L coordinate; 0 H coordinate, 255 S coordinate and 110 L coordinate; 71 H coordinate, 255 S coordinate and 123 L coordinate; 111 H coordinate, 255 S coordinate and 123 L coordinate.

[0093] Alternatively or additionally, steps i. to iv. can be repeated for the retained fractions of the seeds using a second backlight (which may have an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172), wherein steps i. to iv. can be repeated for the ejected fractions of the seeds using a first backlight (which may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110). Alternatively or additionally, in the above method using a first backlight (which may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110), steps i. to iv. can be repeated on the ejected fractions of seeds (e.g., mainly dark blue and light blue seeds) using the first backlight (which may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110), wherein steps i. to iv. can be repeated on the ejected fractions of seeds from the repeated ejected fractions of seeds (e.g., mostly dark blue and light blue seeds) using a third backlight (which may have an H coordinate of 20, an S coordinate of 255, and an L coordinate of 150; or an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110; or an H coordinate of 111, an S coordinate of 255, and an L coordinate of 123).

[0094] The method may further include repeating steps i. to iv. using the retained or ejected fractions of the seeds from the ejected fractions of the seeds used in the repeated steps i. to iv., wherein this further repetition uses a fourth backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The fourth backlight may be different from the first backlight and / or the second backlight. The fourth backlight may have at least one color in the HSL color space selected from the group consisting of: H coordinate 20, S coordinate 255 and L coordinate 150; H coordinate 0, S coordinate 255 and L coordinate 110; H coordinate 71, S coordinate 255 and L coordinate 123; H coordinate 340, S coordinate 192 and L coordinate 172; H coordinate 111, S coordinate 255 and L coordinate 123.

[0095] Fractions of sprayed and / or retained seeds, including colorless grain seeds (specifically composed of them), can be added to repetitions of steps i. to iv. using the retained fractions of seeds. In practice, for example, in methods aimed at obtaining purer colorless seeds (e.g., male-sterile seeds in a colorless seed enrichment stream), retained seeds obtained after blowing out colored (e.g., blue aleurone layer) seeds can be used for further blowing out colored (e.g., blue aleurone layer) seeds to maximize the purity of colorless seeds. Similarly, retained seeds obtained after blowing out light and dark (e.g., light blue and dark blue aleurone layer) seeds in a colored seed stream (e.g., a 1n / 2n blue aleurone layer retention stream) can also be used or added to the seeds used in the colorless (e.g., male-sterile) seed sorting stream to maximize the yield of colorless (e.g., male-sterile) seeds, of course, this is the case when handling large quantities of seeds in large-scale field production.

[0096] The retention of the seed for steps i. to iv. can be repeated at least twice.

[0097] In this application, whenever a specific backlight setting in the HSL color space is referred to herein by way of a specific value of H, S, or L backlight setting (such as a specific HSL setting illustrated by example or a specific HSL setting mentioned in the specification and / or drawings), the specific HSL setting may differ from the given specific value by + / 3, + / - 2, or + / - 1 for H values ​​(on a 0-360 scale), by + / - 15, + / - 10, or + / - 5 for S values ​​(on a 0-255 scale), or by + / - 5, + / - 4, + / 3, + / - 2, or + / - 1 for L values ​​(on a 0-255 scale). Therefore, an H value of 20 can be replaced with an H value of 17 or 23, or any value between 17 and 23; an S value of 192 can be replaced with an S value of 177 or 207, or any value between 177 and 207; and an L value of 123 can be replaced with an L value of 118 or 128, or any value between 118 and 128. Furthermore, these values ​​may need to be adjusted based on the quality of the backlight, camera, etc., used to achieve similar sorting efficiency before using specific HSL backlight settings on another seed sorting machine. Additionally, in some cases, alternative HLS backlights can be used in a specific step (such as in…). Figures 4 to 9 (Those shown in the description).

[0098] In the sorting steps of a seed stream (colorless seed stream) to improve the purity of colorless seeds (such as male-sterile seeds lacking a blue aleurone layer), a first backlight can be used in the first sorting step to retain colorless seeds and blow out colored seeds (such as using the DB_LB blowing method described herein, employing HSL backlights H340 / S192 / L172 or H71 / S255 / L123, to maximize the yield of the sprayed colored (such as blue aleurone layer) seeds, e.g. Figures 4 to 7 (As shown in the left part), or using the DB_LB blowing method described herein, with HSL backlighting H0 / S255 / L110, to maximize the yield of colorless seeds (such as male-sterile seeds without a blue aleurone layer) in the retained fraction (e.g. Figure 8 and Figure 9 (As shown in the left part). In the second sorting step, in the colorless seed stream (such as the male-sterile seed enrichment stream) using the seeds retained using the first backlight in the first sorting step, when the first backlight has HSL settings H340 / S192 / L172 or H71 / S255 / L123, the first backlight can be reused (e.g., repeated 1, 2, or 3 times, or 1-3 times) to purify the colorless seeds (such as male-sterile seeds) and blow out the colored seeds (such as dark blue and light blue seeds) (e.g., using the same or alternative first backlight as in the first sorting step, such as by using the DB_LB blowing method described herein). Figures 4 to 7 (As shown in the left side section). Alternatively, in the second sorting step, in the colorless seed stream using the retained seeds obtained in the first sorting step with the first backlight, when the first backlight has an HSL setting of H0 / S255 / L111, a second backlight can be used to purify colorless seeds (such as male-sterile seeds) and blow out colored seeds (such as dark blue and light blue seeds), as described in the second sorting step using the DB_LB blowing method described herein, with a second backlight HSL setting of H340 / S192 / L172 or H71 / S255 / L123, as... Figure 8 and Figure 9 As shown in the left-hand section.

[0099] In the preceding text, the colored seeds ejected / blown from the first sorting step using the first backlight can be used in the seed stream (colored seed stream) to improve the purity of the colored seeds (such as blue or light blue (1n / 2n retention system)). Therefore, the first sorting step in the colorless seed stream is also the first sorting step in the colored seed stream, using the same first backlight as in the colorless seed stream (however, the colored seed stream uses ejected seeds instead of retained seeds). In the second sorting step of the colored seed stream, a second backlight is used to deplete darker (e.g., 3n dark blue) seeds from the colored seeds (e.g., blue, fertile (maintainer) seeds) ejected in the first sorting step using a first backlight, in order to purify lighter-colored seeds (e.g., 1n / 2n blue seeds) by blowing out the darker seeds (e.g., 3n dark blue seeds). The second backlight has HSL settings H20 / S255 / L150, H111 / S255 / L123, or H0 / S255 / L110 (e.g., using the deep blue (DB) blowing method described herein, which can be achieved using the low L or low L+S method described herein). Figures 4 to 8 (as shown in the right-hand portion), or the second backlight in the second sorting step has an HSL setting H0 / S255 / L110 (such as by using the DB_LB blowing method described herein, employing this second backlight (e.g.) Figure 9 (As shown in the right-hand section).

[0100] Furthermore, in the colored seed stream, a third backlight can be used for a third sorting step to: i) blow out light-colored (such as light blue) seeds (such as by using the LB blowing method described herein, employing a third backlight with HSL settings H0 / S255 / L111, H71 / S255 / L123 or H20 / S255 / L150, as...). Figure 4 and Figure 5 (As shown on the right); or a third backlight with HSL settings H0 / S255 / L111 or H71 / S255 / L123 can be used, such as Figure 8 (as shown in the right-hand section), or ii) blow out light and dark seeds (such as blue and dark blue seeds, such as by using the LB_DB blowing method described herein, employing a third backlight with HSL settings H340 / S192 / L172 or H71 / S255 / L123, as shown in the right-hand section). Figure 6 (As shown in the right-hand section); or by using the LB_DB blowing method described herein, a third backlight with HSL settings H0 / S255 / L110 or H71 / S255 / L123 (such as...) is employed. Figure 7(As shown in the right part)), or iii) blow out dark (such as 3n dark blue) seeds, such as by using the DB blowing method described herein (which can use the low L or low L+S method described herein), employing a third backlight (such as) with HSL settings H20 / S255 / L150, H0 / S255 / L110 or H111 / S255 / L123 Figure 9 (As shown in the right-hand section). The above also includes any backlight having an H coordinate that differs from the aforementioned H coordinate by + / - 3 (within the 0-360 scale), any S coordinate that differs from the aforementioned S coordinate by + / - 15 (within the 0-255 scale), or any L coordinate that differs from the aforementioned L coordinate by + / - 5 (within the 0-255 scale).

[0101] In one embodiment of the above method, in addition to the preferred backlight HSL setting described above, the L-object setting or L and S-object setting for the object to be ejected is also modified from those settings provided by the sorting device, for example by reducing L. min And L max Adjust to a certain number of seeds to be blown out, or lower L. min and S min and L max or L max and S max Adjust to a certain number of seeds to be blown out (e.g., H, S, and L object settings as described in this article specifically for the low L (brightness) or low L+S method for removing dark blue seeds, or as described in this article specifically for the light blue (LB) blowing method for retaining dark and colorless seeds).

[0102] Furthermore, in the aforementioned schemes, such as in Figures 4 to 9In addition to sorting using adjusted backlight HSL settings (with or without reduced / adjusted L or L and S object settings) according to the methods described herein, any other sorting methods or tools that contribute to improvements to achieve the following objectives: maximizing the yield of colorless grain seeds; maximizing the yield of colored grain seeds; maximizing the purity of colorless grain seeds, specifically by removing impurities such as colored seeds, weed seeds, disease-infected seeds (such as those infected with ergot), and / or other plant parts; maximizing the purity of colored grain seeds; reducing the number of colored seeds among colorless seeds; reducing the number of colorless seeds among colored seeds; maximizing the purity of colorless seeds while minimizing colorless seed loss; maximizing the purity of colored seeds while minimizing colored seed loss; or any combination thereof. In this regard, in addition to the methods described herein, the removal or retention of deep blue 3n aleurone layer seeds, light blue 1n / 2n blue aleurone layer seeds, or colorless "white" seeds can be accomplished 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 the use of trained neural networks to improve sorting efficiency. Any other (color) sorting methods or other sorting machines (e.g., Cimbria SEA.IQ PLUS or SMART sorters from AnySort), such as methods or sorters using artificial intelligence (with or without adjustable backlight color), can be used in conjunction with the method of the present invention (before, during, or after) to blow any (remaining) colorless "white" seeds from the blue / colored seed stream. In one embodiment, such other sorting methods can be performed on or by the same sorting equipment as the method of the present invention.

[0103] At least step iii., and optionally at least one of steps ii. and iv., may specifically be at least one of computer implementation and computer control. As used herein, the term "computer implementation" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited 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 include at least one controller configured to perform at least one method step of the method according to the invention. The computer-implemented method step may be performed entirely automatically, specifically without user interaction. As used herein, the term "computer control" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a method or method step involving control by at least one computer and / or at least one computer network. Specifically, computer-controlled method steps may involve using at least one computer and / or computer network, which may include at least one controller configured to control at least one device (such as a sorting station and / or any part thereof) to perform a specific task or function. For example, at least one computer and / or computer network may include at least one controller configured to control a camera to capture images of backlit seeds in a seed stream. Alternatively or additionally, at least one computer and / or computer network may include at least one controller configured to control a sorting station (specifically, an ejector) to eject seeds identified as to be sorted from the seed stream.

[0104] In step iii, the automatic identification of seeds to be sorted from the seed stream from the image captured in step ii can be performed by a trained artificial neural network (ANN). As used herein, the term "artificial neural network," also known as "neural network," is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, a mathematical model comprising multiple units or nodes connected by one or more edges. Each unit or node may receive at least one input signal from one or more connected units or nodes and may be configured to process the at least one input signal and forward at least one output signal to one or more connected units or nodes. The input and / or output signals may be real numbers. The output signal of each unit or node may be calculated by at least one mathematical function (specifically at least one nonlinear function) taking into account the sum of the input signals. The mathematical function of the unit or node may be the activation function of the unit or node. The strength of the input signal at each connection can be determined by using adjustable weights, which can be adjusted during the learning or training process. These units or nodes may be aggregated into two or more layers. Input signals may be provided to the input layer and may be forwarded to the output layer. Alternatively, an artificial neural network may include one or more hidden layers between the input layer and the output layer.

[0105] The method may further include at least one training step. In this training step, labeled images can be used to train the ANN. The labeled images may include genotyping data and / or provide information about seeds with a normal (colorless) aleurone layer and cereal seeds containing a blue aleurone layer. Therefore, the labeled images can provide benchmark real data for the training step.

[0106] A trained ANN can be trained using a record of training data. This record can include training input data and corresponding training output data. The training output data of the training data record can be used to predict the results produced by the ANN when the same training input data from the same training data record is fed into the ANN. The deviation between this expected result and the actual result produced by the ANN can be observed and evaluated using a "loss function." This loss function can be used as feedback to adjust the ANN parameters. For example, the parameters can be tuned to minimize the value of the loss function when all training input data is fed into the ANN and the results are compared with the corresponding training output data. The result of this training might be that, given a relatively small number of training data records as a "benchmark," the ANN can perform well on a much larger number of input data records. Therefore, an ANN can include at least one algorithm and model parameters. The parameters of the ANN can be adjusted during the training process.

[0107] The method may further include at least one second spectral seed sorting step. As used herein, the term "spectral seed sorting step" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, the process of sorting seeds using a spectroscopic device. Specifically, the second spectral seed sorting step may include the use of at least one spectrometer device, such as a device configured to acquire at least one optical property or optically measurable property as a function of wavelength. In particular, the spectrometer device used in the second spectral seed sorting step may be, or may include, a device configured to record signal intensity relative to a corresponding wavelength of a spectrum or its division (such as a wavelength range). The second spectral seed sorting step may include one or more of infrared spectral analysis (specifically near-infrared spectral analysis), UV spectral analysis, spectral analysis using X-rays, and / or spectral analysis using Raman scattering. The second spectral seed sorting step (such as near-infrared spectral analysis) may be performed simultaneously with or sequentially with steps i. to iv. For example, the second spectral seed sorting step may be performed after steps i. to iv., for example, as a subsequent sorting round of steps i. to iv. Alternatively, the spectral seed sorting step can be performed concurrently with steps i. to iii. in time, wherein the ejection in step iv. can be performed based on the results of step iii. and / or the second spectral sorting step.

[0108] The second spectral seed sorting step may include determining at least one spectral piece of information about the seeds in the seed stream, which differs from the information determined in the first visible light sorting step described above (specifically in method steps i. to iii.). This spectral information may include at least one of transmission, absorption, reflection, and emission of the seeds in the seed stream relative to a corresponding wavelength of the spectrum or its division (such as a wavelength range). The second spectral seed sorting step may include determining at least one of transmission, absorption, reflection, and emission in the infrared spectral range (e.g., in the wavelength range of 760 nm to 1000 μm), specifically in the near-infrared spectral range (e.g., in the wavelength range of 760 nm to 1.5 μm). This at least one spectral piece of information can be used to automatically identify seeds to be sorted from the seed stream. For example, this spectral information can be used to determine one or more of the presence, absence, and concentration of a specific compound (e.g., an organic colorant, such as anthocyanin) in the seed. Methods for identifying anthocyanin content using NIR spectroscopy are known to those skilled in the art, for example, as described in Stuppner et al. (2020, Sensors 20(17), 4983 (https: / / doi.org / 10.3390 / s20174983), Chen et al. (2015, FoodChemistry 172, 788-793 (https: / / doi.org / 10.1016 / j.foodchem.2014.09.119), or WO2023 / 088892 A2. Seeds in the seed stream for the detection of one or more of the presence, absence, and concentration of a specific compound can be sprayed using a sprayer at a sorting station.

[0109] The second spectral seed sorting step may further include a trained artificial neural network that automatically identifies seeds to be sorted from the seed stream based on the at least one spectral piece of information. The method may further include at least one training step. In the training step, the ANN can be trained using tagged spectral information. This tagged spectral information may include genotyping data and / or provide information about seeds with a normal (colorless) aleurone layer and cereal seeds containing a blue aleurone layer. Therefore, the tagged spectral information can provide benchmark real data for the training step.

[0110] The method may further include at least one control analysis step. This control analysis step may include using a multispectral imaging device with a black background (such as a black conveyor belt). Specifically, the multispectral imaging device may have a black conveyor belt. Specifically, in terms of visual inspection, a black conveyor belt can provide the most accurate control results.

[0111] In another aspect of the invention, a sorting device for sorting grain seeds is disclosed. As used herein, the term "sorting device" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a device configured to perform at least one sorting function. Specifically, the sorting device may be configured to distinguish and separate seeds to be sorted from other seeds in a seed stream. As will be further detailed below, the sorting device includes at least one seed feeder for supplying the seed stream. The sorting device may be or may include an optical sorting device. Specifically, the sorting device may include at least one optical system configured, in conjunction with image processing software, to identify seeds to be sorted in the seed stream. For example, a camera and backlighting device may be part of the optical system of the sorting device. The sorting device may further include at least one separation system for performing the separation of the seeds to be sorted from other seeds in the seed stream. For example, an injector may be part of the separation system.

[0112] The sorting equipment includes:

[0113] I. At least one seed feeder for supplying a seed stream to at least one sorting station; and

[0114] II. At least one sorting station, including at least one backlighting device for backlighting the seeds of the seed stream, the sorting station further including at least one camera for capturing at least one image of the backlit seeds, and the sorting station further including at least one ejector for ejecting the seeds from the seed stream.

[0115] The sorting device is configured to perform the method according to the present invention (e.g., according to any of the embodiments disclosed above and / or according to any of the embodiments disclosed in further detail below). Therefore, for the definition of terms and / or the description of possible embodiments, refer to the description of the method for sorting grain seeds above.

[0116] As used herein, the term "seed feeder" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, a device configured to supply a stream of seeds. Specifically, a seed feeder may include at least one feed hopper configured to receive a plurality of seeds and supply them in a controlled manner to at least one chute. The seed feeder may further include at least one vibratory feeder configured to apply vibration to the feed hopper, such that seeds included therein may exit the feed hopper and enter the at least one chute. The seed feeder may be configured to supply a plurality of seeds arranged in a regular pattern. The seed feeder may be configured to supply a plurality of seeds arranged individually in a row or line.

[0117] The sorting device may further include at least one controller. As used herein, the term "controller" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, any logic circuit configured to perform basic operations of a computer or system, and / or generally refers to a device configured to perform computational or logical operations. In particular, the controller may be configured to process basic instructions that drive the computer or system. As an example, the controller may include at least one arithmetic logic unit (ALU), at least one floating-point unit (FPU) (such as a math coprocessor or a digital coprocessor), multiple registers (specifically configured to provide operands to the ALU and store the results of operations), and memory (such as L1 and L2 cache memories). In particular, the controller may be a multi-core processor. Specifically, the controller may be or may include a central processing unit (CPU). Additionally or alternatively, the controller may be or may include a microprocessor; therefore, specifically, the elements of the controller may be contained within a single integrated circuit (IC) chip. Additionally or alternatively, the controller may be or may include 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 units (TPUs) and / or one or more chips, such as dedicated machine learning optimization chips. Specifically, the controller may be configured (e.g., via software programming) to perform one or more operations. The controller may be configured to perform at least step iii of the method. Optionally, the controller may be further configured to control at least one of steps ii and iv of the method.

[0118] The sorting device may further include at least one target sluice and at least one sorting sluice. As used herein, the term "target sluice" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a seed channel dedicated to seeds that are not to be sorted. As used herein, the term "sorting sluice" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a seed channel dedicated to seeds to be sorted. The target sluice and the sorting sluice may specifically be different from each other. Seeds ejected by the ejector may be collected by the sorting sluice. The remaining seed stream may be collected by the target sluice. When seeds are identified as seeds to be sorted, the sorting station (specifically the ejector) may be configured to separate these seeds from the seed stream into the sorting sluice. The sorting device may be configured such that seeds in the seed stream that are not identified as seeds to be sorted may pass through the sorting station toward the target sluice.

[0119] The sorting station can be configured to allow the seed stream to pass between the backlight device and the camera.

[0120] A sorting station may include at least two cameras. The cameras may be configured to capture images of backlit seeds from different angles in space. The sorting station may further include at least two backlighting devices. Each backlighting device may be assigned to one of the at least two cameras. The cameras may be configured to capture images of the backlit seeds from opposite directions. For example, the sorting station may include at least one first backlighting device assigned to at least one first camera. The first backlighting device may be arranged on a first side relative to the seed stream. The first camera may be arranged on a second side relative to the seed stream, opposite the first side. Further, the sorting station may include at least one second backlighting device and at least one second camera. The second backlighting device may be arranged on the second side, and the second camera may be arranged on the first side.

[0121] As used herein, the term "ejector" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, any device configured to eject seeds. An ejector may specifically be or may include at least one of mechanical and pneumatic devices. An ejector may include at least one of a mechanical ejector having at least one mechanical actuator and a pneumatic ejector having at least one nozzle for directing an air jet to seeds to be sorted from a seed stream. For example, an ejector may include at least one pneumatic ejector configured to eject seeds to be sorted from a seed stream by using compressed air directed via one or more nozzles to separate the seeds to be sorted from the seed stream.

[0122] In another aspect of the invention, a computer program is disclosed that includes instructions, when executed by a sorting device according to the invention (such as according to any embodiment described above and / or according to any embodiment further detailed below), particularly by a controller of the sorting device, to cause the sorting device to perform at least step iii. of the method according to the invention (such as according to any embodiment described above and / or according to any embodiment further detailed below), and optionally at least one of steps ii. and iv. of the method.

[0123] Therefore, specifically, one, more, or even all of the method steps i to iv as described above can be executed and / or controlled by using a computer or computer network, preferably by using a computer program.

[0124] In another aspect of the invention, a computer-readable storage medium is disclosed, specifically a non-transient computer-readable medium comprising instructions that, when executed by a sorting device according to the invention (such as according to any embodiment described above and / or according to any embodiment further detailed below), particularly by a controller of the sorting device, cause the sorting device to perform at least step iii. of the method according to the invention (such as according to any embodiment described above and / or according to any embodiment further detailed below), and optionally at least one of steps ii. and iv. of the method.

[0125] As used herein, the term "computer-readable data medium" can specifically refer to a non-transitory data storage device, such as a hardware storage medium on which computer-executable instructions are stored. Specifically, a computer-readable storage medium can be or may include storage media such as random access memory (RAM) and / or read-only memory (ROM). A computer-readable storage medium can be or may include at least one computer-readable data carrier.

[0126] This document further discloses and proposes a computer program product having program code means so that, when the program is executed on a computer or computer network, it performs and / or controls the methods according to the invention in one or more embodiments disclosed herein. Specifically, the program code means may be stored on a computer-readable data carrier and / or a computer-readable storage medium.

[0127] This document further discloses and proposes a data carrier having a data structure stored thereon, which, after being loaded into a computer or computer network (such as into the working memory or main memory of the computer or computer network), can execute and / or control the execution of methods according to one or more embodiments disclosed herein.

[0128] This document further discloses and proposes a computer program product having program code means stored on a machine-readable medium for executing and / or controlling the execution of methods according to one or more embodiments disclosed herein when the program is executed on a computer or computer network. As used herein, a computer program product refers to a program that is a tradable product. The product can generally exist in any format, such as in paper format, or on a computer-readable data carrier and / or computer-readable storage medium. Specifically, the computer program product can be distributed via a data network.

[0129] Finally, this document discloses and proposes a modulated data signal containing computer system or computer network readable instructions for performing and / or controlling methods according to one or more embodiments disclosed herein.

[0130] Referring to the computer implementation aspects of the present invention, one or more method steps, or even all method steps, of the methods disclosed herein according to one or more embodiments can be performed and / or controlled using a computer or computer network. Therefore, generally, any method step, including data provision and / or operation, can be performed using a computer or computer network. Typically, these method steps can include any method steps except for those that typically require manual work, such as providing samples and / or performing certain aspects of actual measurements.

[0131] In another aspect of the invention, the use of a sorting device according to the invention (such as according to any of the embodiments described above and / or according to any of the embodiments further detailed below) for the purpose of selecting from the group consisting of: sorting colored seeds from a seed mixture containing colorless seeds and colored seeds; sorting dark seeds (with or without colorless seeds) from a seed mixture containing dark seeds and lighter-colored seeds; sorting lighter-colored seeds from a seed mixture containing dark seeds and lighter-colored seeds, which may contain colorless seeds; sorting dark seeds and lighter-colored seeds from a seed mixture containing dark seeds, lighter-colored seeds and colorless seeds; sorting from a seed mixture containing colorless grain seeds and colored grains. Colored grain seeds are sorted from a mixture of grain seeds; colored grain seeds containing a blue aleurone layer are sorted from a mixture of grain seeds containing colorless grain seeds and grain seeds containing a blue aleurone layer; 3n dark blue aleurone layer seeds are sorted from a seed mixture containing 1n, 2n, and 3n blue aleurone layer seeds (with or without colorless seeds); lighter 1n and 2n blue aleurone layer seeds (with or without colorless seeds) are sorted from a seed mixture containing 1n, 2n, and 3n blue aleurone layer seeds; lighter 1n, 2n, and darker 3n blue aleurone layer seeds are sorted from a seed mixture containing 1n, 2n, and 3n blue aleurone layer seeds and colorless seeds; using a color space with 70... Using at least one backlight with an H coordinate of ≤ H ≤ 150, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer; using at least one backlight with an H coordinate of 25 ≤ H ≤ 70 in the HSL color space, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer; using at least one backlight with an H coordinate of 0 ≤ H < 25 or H ≥ 310 in the HSL color space, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer.

[0132] Specifically, a sorting device according to the invention, used for separating colored grain seeds containing a blue aleurone layer from a mixture of colorless grain seeds (without a blue aleurone layer) and grain seeds containing a blue aleurone layer using at least one backlight with an H coordinate of 70 ≤ H ≤ 150 in the HSL color space, allows for the high-purity sorting of colorless grain seeds. Specifically, during sorting, white seeds can be retained, and blue seeds can be blown out and used, for example, when it is necessary to retain seed stock and / or to seek higher anthocyanin colorant content in the seeds. Furthermore, 3n bla seeds can be removed from 1n and 2n blue seeds. Alternatively or additionally, any blue seeds can be removed from white seeds by blowing out light blue and dark blue seeds.

[0133] Using the sorting device according to the invention for separating colored grain seeds containing a blue aleurone layer from a mixture of colorless grain seeds and grain seeds containing a blue aleurone layer using at least one backlight with an H coordinate of 25 ≤ H ≤ 70 in the HSL color space, efficient sorting can be achieved, i.e., sorting colorless grain seeds with good purity and high product yield.

[0134] Using the sorting device according to the invention for separating colored grain seeds containing a blue aleurone layer from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer using at least one backlight with an H coordinate having 0 ≤ H < 25 or H ≥ 310 in the HSL color space, very good purity and good product yield can be achieved in the colorless grain seeds for seed streams of grain seeds with a deep blue phenotype containing a blue aleurone layer and colored seeds (especially in seeds with a 3n blue aleurone layer).

[0135] The method and apparatus for sorting grain seeds according to the present invention can provide numerous advantages over known methods and apparatus. Specifically, by using colored backlighting, the method for sorting grain seeds can improve sorting purity and sorting yield when sorting grain seeds containing colorless grain seeds and seeds containing a blue aleurone layer.

[0136] This method can include improved seed sorting based on color differences in mature plant seeds, particularly wheat seeds. Different colors of mature plant seeds can be used to sort a particular type of seed from other types, such as seeds containing a color gene as a selectable marker to sort another gene closely linked to the color gene. This method can be particularly useful when a different color of seed implies a higher anthocyanin content. Additionally or alternatively, this method can be specifically used in the field of hybridization breeding, where the maternal and paternal plants have seeds of different colors, and the difference in seed color is used to remove self-pollinated paternal seeds from hybrid seeds (e.g., when self-pollinated paternal seeds are harvested together with hybrid seeds, such as when a certain proportion of paternal plants are intercropped among most maternal plants to produce hybrid seeds). For example, in some cases, blue seeds may not be permitted in a particular grain seed being sold, while blue seeds with higher anthocyanin content may be present in the grain. This method can reliably sort seeds even when seed color intensity depends on the copy number of a color locus (such as the BLA locus), and sorting different color intensities from colorless seeds is challenging. This method can even be used to isolate seeds whose aleurone layers contain one, two, or three copies of color loci (such as the BLA locus).

[0137] By using colored backlighting, this method can specifically provide accurate color sorting of grain kernels containing the BLA locus. In particular, the ability to sort non-BLA grades to extremely high purity levels of 99% or higher, or 99.7% or higher, may depend on the specific background light setting used. In this method, using color values ​​in the HL color space with H coordinates of H ≤ 150 or H ≥ 310 and L coordinates of 0.10 ≤ L ≤ 0.80, such as yellow, red, and / or green backlighting, sorting purity can exceed the 99% threshold, specifically exceeding 99.7%, more specifically obtained through repeated blowing, particularly when sorting two to three times. Furthermore, using different colored backlights may affect the amount of loss occurring during sorting into different colored seed grades. Furthermore, in this method, backlight color values ​​with H coordinates of H ≤ 150 or H ≥ 310 and L coordinates of 0.10 ≤ L ≤ 0.80 in the HSL color space are used. For example, for yellow, red, and / or green backlights, when achieving a purity of over 99% or 99.7%, the loss of colorless (“white” seeds, which are ultimately “white” seeds that enter the blown fraction, may be 4-45%, 4-40%, 4-35%, 4-30%, 4-25%, 4-20%, 4-15%, or 4-10%. Using the multiple-blow method of the present invention (without recovering white seeds from the maintainer (blue) seed stream), a mixture of (fertile) blue aleurone layer and (male-sterile) colorless seeds (produced in favorable growing conditions in German field trials) was sorted according to 11 different blue aleurone layer wheat varieties. An average of 11.5% (by weight) white seed loss was observed across these 11 varieties, with "white" seed losses ranging from 4% to 33%, and the average purity of "white" seeds across these 11 different varieties was 99.6% (ranging from 99.2% to 99.8%). Therefore, using the multiple-blow method of the present invention, "white" seed losses of 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% can be obtained when achieving a white seed purity of at least 99% or at least 99.7%. Furthermore, using the method of the present invention employing multiple blowouts (without recovering blue seeds from the male-sterile (“white”) seed stream), when sorting a mixture of (fertile) blue aleurone layer and (male-sterile) colorless seeds produced in a single field trial in a favorable growing environment in Germany from four different varieties, the average loss (by weight) of blue aleurone layer seeds in the (blue seed) maintainer seed stream was 10% (ranging from 7% to 14% across the four genotypes), and the purity of blue aleurone layer seeds was at least 95% on average across these four genotypes. Therefore, different backlighting can have a significant impact on sorting efficiency.Green backlighting, such as any green backlighting with HSL values ​​as specifically described herein, yielded optimal sorting results for blowing blue seeds from a mixture of blue and white seeds across all tested genotypes. Red backlighting, such as any red backlighting with HSL values ​​as specifically described herein (including H=0 (H=360)), achieved the highest product yield for white seeds and was best used to separate blue and white fractions in the first sorting step to obtain blue seeds, followed by the removal of dark blue seeds (3n BLA seeds) from lighter blue seeds (1n and 2n BLA seeds) in a second or third sorting step, wherein the second or third sorting step preferably used a yellow or orange backlighting color, such as yellow, orange, or red backlighting with HSL values ​​as specifically described herein. Furthermore, yellow or orange backlighting, such as yellow or orange backlighting with HSL values ​​as specifically described herein, provided optimal results for sorting dark blue seeds with the 3n BLA locus in the aleurone layer from lighter blue seed (1n and 2n BLA) samples.

[0138] The method for sorting grain seeds can specifically provide better results when sorting blue seeds compared to sorting white seeds. This method can be used to obtain batches of high-purity white seeds, especially when using at least two sorting runs. This can be achieved, for example, on standard large-scale seed sorting equipment by using multiple sluices operating in tandem within the same machine, employing multiple sorting runs. Therefore, generally speaking, for genotypes that are more difficult to sort, more blow-out runs can improve sorting purity. Multispectral imaging equipment (such as through the use of Videometer®) can be used as a quality control system to check the purity of sorted seeds. Additionally or alternatively, genotyping data can be used for quality checks. Multispectral imaging equipment can typically be used to distinguish 3n blue aleurone layer haplotype seeds (dark blue seeds) from “normal” blue 2n seeds or light blue 1n blue aleurone layer haplotype seeds. 3n blue aleurone layer haplotype seeds may be much darker in color than 2n blue aleurone layer haplotype seeds. The amount of dark blue 3n seeds can be reduced from the blue seed batch by sorting out the darkest blue seeds (see above), for example by reducing the L of the scheme. min (Object settings) Value reduced to 0 or reduced to 0 ≤ L min ≤ 60, specifically reduced to 0 ≤ L min ≤ 50, more specifically reduced to 0 ≤ L min ≤ 40, 0 ≤ L min ≤ 30, 0 ≤ L min ≤ 20 or 0 ≤ L min ≤ 10, or even more specifically reduced to 0 ≤ L min ≤ 30 (depending on the color depth of the seed batch), and Lmax The (object) value decreases to 127 from 85 or 123 from 81, or L max The value is 78 ≤ L max ≤ 127, depending on the percentage or amount of 3n seeds to be removed, and depending on the color depth of the seed batch, specifically using the method described herein for removing / blowing out blue aleurone layers of seeds (including...). Figures 4 to 9 Any colored backlight (as mentioned in the description), such as any colored backlight 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 backlight having an H coordinate of 20 ≤ H ≤ 70, an S coordinate of 0.25 ≤ S ≤ 1, and an L coordinate of 0.35 ≤ L ≤ 0.78. In particular, such a backlight has an 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 (for blowing out darker (e.g., dark blue) seeds), a white LED backlight or a backlight having an L coordinate of L = 255 (white) can also be used. In this method, L... min The (object) value is best set to 0 (the safest choice, as it works well and is independent of seed color depth), but it can be increased to 50 for dark seed batches and to 60 for light seed batches without significantly affecting the percentage of dark blue seeds blown out. Similarly, in this (low L) method of blowing out 3n blue aleurone layer seeds from a blue seed batch, L... max It depends on the amount of 3n blue aleurone seeds to be removed. For example, on the L scale from 0 to 255, the optimal L for blowing out 10% deep blue seeds is... max The values ​​are 77 ≤ L ≤ 87, 79 ≤ L ≤ 95 for 15% of seeds blown out, 81 ≤ L ≤ 100 for 20% of seeds blown out, 84 ≤ L ≤ 103 for 30% of seeds blown out, 87 ≤ L ≤ 107 for 40% of seeds blown out, and 91 ≤ L ≤ 110 for 50% of seeds blown out. In one embodiment, L here... max The (object) value is set between 85 and 89, such as 85, 86, 87, 88, or 89 (on a scale of 0 to 255). In this low L (or brightness) method, the hue object setting 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 including the range of H ≥ 354 and H ≤ 54 (on a scale of 0-360 H).

[0139] In one embodiment of the invention, the amount of darker grain seeds (such as dark blue 3n grain seeds) can also be reduced from lighter grain seeds (such as lighter blue 1n / 2n seeds) by sorting out the darkest seeds (such as dark blue 3n seeds), for example by reducing the L of the scheme. min (Object settings) value. L min It can be set to 0, or set within the range of 0 to 60, specifically 0 to 50, more specifically 0 to 40, 0 to 30, 0 to 20, or 0 to 10, or even more specifically 0 to 30 (including the endpoints, depending on the color depth of the seed batch), and L max The (object) value is set in the range of 78 to 120, 81 to 123, or 78 to 127 (on a 0-255 scale, depending on the percentage or amount of 3n seeds to be removed, and the color depth of the seed batch).

[0140] The aforementioned method for reducing the amount of dark blue 3n seeds in a batch of blue seeds (low L) can be used, for example, in a second or third sorting step after the initial sorting of blue seeds (retaining white seeds), using the fraction blown out in the earlier sorting step as described above (e.g., in...). Figures 4 to 9 In the schematic DB blue blowing step, the adjusted backlight HSL setting of the present invention is used.

[0141] In this context, it is useful to note that in segregating ears of cereal plants with a hybrid system based on the blue aleurone layer locus on the same chromosome (arm) as the fertility restorer gene (which, in plants with a male-sterile gene, results in male sterility in the absence of the restorer gene), the percentage of 3n (double blue, dimorphic) deep blue seeds in the blue seed fraction is approximately 10-20%. If this blue seed fraction is used for seed propagation without depleting the dimorphic deep blue seeds, the percentage of deep blue 3n seeds increases to over 30% in the first generation, over 60% in the second generation, and over 90% in the third generation. Without depleting the dimorphic deep blue seeds, the percentage of "white" seeds obtained in maintainer line breeding decreases from approximately 66 / 34 (white / blue) to 60 / 40 in the next breeding cycle, to 45 / 55 in subsequent breeding cycles, and to 31 / 69 in the next breeding cycle (assuming no removal of 3n blue aleurone layer seeds). Therefore, the amount of male-sterile females obtained from a given area is reduced, thus increasing production costs. Therefore, it is recommended to reduce the amount of dark blue 3n seeds in the blue seed fraction through a specific seed sorting step, in which either the darkest seeds in the blue seed fraction are blown out, or alternatively, only the light blue seeds are blown out and only the light blue fraction is used for seed propagation. The sorting step can also be used to remove (darker) colored seeds from seed batches containing both (darker) colored and colorless or light-colored seeds.

[0142] As described above, step iii. includes automatically identifying the seeds to be sorted from the seed stream from the image captured in step ii. In step iii., the seeds to be sorted from the seed stream can be identified by recognizing objects in the image that have predefined color coordinates provided by the sorting device (on the captured image), specifically objects with color coordinates within a predefined range in the HSL color space. In one embodiment of the invention, the method may include the sorting method described above, wherein at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, and wherein the predefined object HSL settings provided by the sorting device for the seeds to be sorted may be intentionally changed or modified to blow out certain seeds or a certain amount of seeds, such as blowing out (a certain amount) (darker) colored seeds from colorless or lighter-colored seeds, or blowing out (a certain amount) dark blue (3n) blue aleurone layer seeds from 2n / 1n blue aleurone layer seeds or colorless "white" seeds, or blowing out light blue (1n / 2n) seeds from dark blue and colorless seeds. In one embodiment, the modification of the predefined object HSL settings may cover the modification of the L object settings, such as L min The object setting is set below the L provided by the sorting device for the (darker) colored seeds to be blown out.min Object settings; or the modifications to the L object settings combined with modifications to the S object settings, such as for (darker) colored seeds to be blown out, S min The object setting is set below the S provided by the sorting device for the (darker) colored seeds to be blown out. min Object settings (where L) max or L max and S max It is set to a value used to blow out a certain percentage of (darker) colored seeds), and the H object setting is any H object range, including 354 ≤ H ≤ 45.

[0143] In one embodiment, the amount of darker seeds (such as dark blue seeds containing 3n blue aleurone layers) can also be reduced from colorless or lighter-colored seeds (such as 1n and / or 2n blue aleurone layer seeds) by sorting out the darkest seeds (see above), for example by reducing the L of the scheme. min and S min (Object settings) value. L min It can be set to 0, or set within the range of 0 to 60, specifically 0 to 50, more specifically 0 to 40, 0 to 30, 0 to 20, or 0 to 10, or even more specifically 0 to 30 (including the endpoints, depending on the color depth of the seed batch), and L max The (object) value is set in the range of 78 to 120, 81 to 123, or 78 to 127 (on a 0-255 scale, depending on the percentage or amount (by weight) of darker seeds (such as 3n blue aleurone layer seeds) to be blown out / removed, and depending on the color depth of the seed batch). min It can be set to 0, or set within the range of 0 to 40, specifically 0 to 30, 0 to 20, 0 to 17, or 0 to 10, or even more specifically within the range of 0 to 30 (including the endpoints, depending on the color depth of the seed batch), and S max The (object) value is set from 60 to 78 or from 60 to 99, depending on the percentage or amount of darker seeds (such as 3n blue aleurone layer seeds) to be removed, and on the color depth of the seed batch (on a 0-255 scale). In this low L+S method, any of the colored backlights described herein can be used to remove / blow out darker seeds, such as 3n blue aleurone layer seeds (including...). Figures 4 to 9The colored backlights mentioned herein, such as any colored backlight having a hue value of 0, 20, 47, 71, 340, or 111, or more specifically, a yellow or orange backlight having an H coordinate of 0 ≤ H ≤ 111 or 20 ≤ H ≤ 70, an S coordinate of 0.25 ≤ S ≤ 1, and an L coordinate of 0.35 ≤ L ≤ 0.78. In particular, such a backlight has an 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 (for blowing darker seeds, such as dark blue seeds), the backlight may also be a white LED backlight or a backlight with an L coordinate of L = 255 (white). In this method, L... min The (object) value is best set to 0, but can be increased to 50 for dark seed batches and to 60 for light seed batches without significantly affecting the percentage of dark blue seeds blown out. Similarly, in this (low L+S) method of blowing out 3n blue aleurone layer seeds from blue seed batches, L... max and S max The amount of 3n blue aleurone seeds to be removed depends on the target amount. In this method, a target percentage (by weight) of dark blue seeds is typically blown out, and the desired percentage is achieved by adjusting the L and S settings (e.g., starting with a test batch of 1000 grams from a larger batch of mixed seeds with approximately 30% dark blue seeds, and targeting the removal of 20% of the dark blue seeds, adjusting the L and S settings within the aforementioned range until approximately 200 grams are blown out—then these L+S settings can be used for the entire seed batch to remove most of the dark blue seeds). Generally, the low L+S method allows for more precise adjustments to the specific amount of darker seeds to be blown out compared to the low L method, because the effect of the low L+S method stems from the interaction of the L and S object settings. The hue object setting 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 in the range of H ≥ 354 and H ≤ 54 (on the 0-360 H scale). The aforementioned (low L+S) method for reducing the amount of dark blue 3n seeds in a batch of blue seeds can be used, for example, in a second or third sorting step after first sorting out the blue seeds (retaining the white seeds), using the fraction blown out in the earlier sorting step as described above (e.g., in...). Figures 4 to 9 In the schematic diagram of the DB blue blowing step, the following is used: Figures 4 to 9 The preferred backlight HSL settings indicated in the description are provided.

[0144] The percentage of deep (3n) blue aleurone seeds in a seed batch can be obtained through genotyping (e.g., copy number analysis of the BLA locus) or through observational experiments. In observational experiments, the amount of plants producing only blue spikes can be measured, from which the average percentage of deep blue seeds in the next generation can be calculated (since it can be expected that a particular plant genotype will have the same or similar paternal inheritance rate when grown in the same environment). Once the percentage of deep blue seeds is estimated, a certain percentage of deep blue seeds can be sorted from the blue seed fraction using the low-L or low-L+S methods described herein. The percentage blown out should not exceed the estimated percentage of 3n blue aleurone seeds (double blue seeds) present in the seed batch.

[0145] Furthermore, the amount of dark seeds (such as blue aleurone layer 3n seeds) and colorless "white" seeds can be reduced from colored (such as blue) seed batches by blowing out lighter-colored seeds (such as lighter blue (1n / 2n aleurone layer) seeds) and retaining the darkest blue and "white" colorless seeds (also referred to herein as the light blue (LB) blowing method). In this LB blowing method, the L of the light blue seeds to be blown out... min The object setting can be: 96 ≤ L min ≤ 110, such as L min It is 105, and S min The object setting can be: 55 ≤ S min ≤ 75, such as 65 ≤ S min ≤ 75, such as S min It is 70, while L max It could be: 138-145, such as L max It is 140, and S max It can be: 95-255, such as S max The value is 120. In one embodiment of this LB blowing method, S min It can be 65 ≤ S min ≤ 75, and L min It can be 97 ≤ L min ≤ 110. In one embodiment of this light blue blowing method, S min The object setting is from 55 to 62 (or 55 ≤ S) min ≤ 62), and S max It is 98, and L min It is from 96 to 102 (or 96 ≤ L) min ≤ 102), and L maxIt is 140 (on the 0-255 scale). This (LB blow-out) method for reducing the amount of dark (such as blue 3n) and "white" seeds in a batch of colored (such as blue) seeds can, for example, first sort out the colored (such as blue) seeds in a first sorting step (retaining white seeds), and then in a second sorting step use the fractions retained in the second sorting step to blow out (3n aleurone layer) dark seeds from the colored seeds (as described above, for example, see...) Figures 4 to 9 Some of the accompanying diagrams schematically illustrate the LB blue blowing process, and in Figures 4 to 9 The preferred backlight HSL setting indicated in the description is used in the third sorting step.

[0146] The hue object setting in this LB blowing method can be any hue range including H ≥ 354 and H ≤ 45, such as H coordinates from H ≥ 345 to H ≤ 45 or from H ≥ 354 to H ≤ 54 (on a 0-360 H scale). Although the low L and of course low L+S methods are preferred, in some cases, this LB blowing method can also be used to blow darker seeds (such as 3n aleurone layer dark blue seeds) from seed batches that are mostly colored (such as mostly blue aleurone layer) as a supplement or alternative to the aforementioned low L or low L+S methods, and can also be used to blow colorless "white" seeds from seed batches that contain mostly colored seeds (such as blue aleurone layer seeds) and some remaining "white" seeds (although its efficiency is mostly lower than the low L or low L+S methods described herein).

[0147] While it would be ideal to blow out all 3n aleurone layer dark blue seeds from maintainer lines / blue seeds in hybrid systems that use the blue aleurone layer as a color marker (located on the same chromosome or chromosome arm as the restorer gene used to restore male sterility in this hybrid system) (which would produce only (fertile and blue) 3n blue aleurone layer offspring), this is only achievable with significant loss of maintainer line seeds due to the overlapping color intensities of 2n and 3nBla seeds. In practice, not all 3n blue seeds are blown out, as the goal is to minimize the loss of maintainer line seeds with 1n / 2n aleurone layers, which can be regenerated to produce male-sterile maternal lines used as maternal lines in hybrid production. The primary objective regarding the removal of 3n aleurone layer seeds is to optimize the efficiency of maintainer line and white seed production and to offset the rapid increase in 3n blue aleurone layer seeds. This means that any blowout of dark blue seeds should result in a higher ratio of 1n / 2n aleurone layer seeds to 3n aleurone layer seeds.

[0148] In one embodiment of the invention, the colored seed to be blown may be a (darker) colored seed whose hue object setting (when the different preferred backlight colors described herein are provided) does not change significantly with other backlight colors (wherein a significant change means that: the H-object setting provided by the sorting device under one backlight color (e.g., for H0 backlight, H-object setting is 180-240) does not overlap with the H-object setting provided by the sorting device under another backlight color (e.g., for H71 backlight, H-object setting is 30-92), or a large portion of the H-object setting range provided by the sorting device for one backlight color is outside the range of 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 blown may not be a dark black seed.

[0149] As another example, the method may include sorting only the 1n and 2n blue aleurone layer seeds from a mixture of 1n, 2n, 3n, and white seeds. By blowing out the light blue seeds, white seeds and the darkest blue 3n bla seeds can be reduced from the blue fraction. The optimal backlight for this purpose may be a red backlight, particularly one with an H coordinate of 0, an S coordinate of 1, and an L coordinate of 0.43, preferably using the modified object setup as described in the LB blowing method above.

[0150] After harvest and before sorting activities, backlighting and / or sorting settings can be optimized individually, for example by performing test sorting to determine the optimal colored backlighting and optimal sorting settings for seeds produced in that environment, depending on the sorting purpose, such as optimizing the amount of colorless white seeds with little or no blue seed contamination, and / or optimizing the purity of colored seeds (such as 1n and 2n BLA seeds) with little or no colorless white seed contamination and / or little or no dark blue (3nBLA seeds).

[0151] Green backlighting can be specifically used to obtain the highest purity of white seeds, especially when there are fewer sorting rounds. It can be considered the most stringent backlighting, but may reduce the amount of white seeds compared to, for example, red backlighting. For a specific seed batch, red backlighting is preferred to obtain a purity higher than 99% or 99.7% while retaining the majority of white seeds in the sorted seed fraction. For a specific genotype, yellow backlighting is preferred to balance purity and quantity, and if the highest purity is not required, yellow backlighting may be the optimal backlighting for retaining the most white seeds across several genotypes compared to red or green backlighting.

[0152] As used herein, the terms “have,” “include,” or “contain,” or any of their grammatical variations, are used in a non-exclusive manner. Thus, these terms can refer either to a situation where no other features exist in the entity described in the context besides those introduced by these terms, or to a situation where one or more other features exist. For example, the statements “A has B,” “A includes B,” and “A contains B” can refer either to a situation where no other elements exist in A besides B (i.e., A consists solely of B), or to a situation where entity A contains one or more other elements besides B (such as element C, elements C and D, or even other elements).

[0153] Furthermore, it should be noted that the terms "at least one," "one or more," or similar expressions indicating a feature or element may appear once or more, but are typically used only once when the corresponding feature or element is introduced. In most cases, the expressions "at least one" or "one or more" are not repeated when referring to the corresponding feature or element, but in fact, the corresponding feature or element may appear once or more.

[0154] Furthermore, as used herein, the terms “preferredly,” “more preferably,” “particularly,” “more particularly,” “specifically,” “more specifically,” or similar terms are used in combination with optional features without limiting the possibility of alternatives. Therefore, the features introduced by these terms are optional features and are not intended to limit the scope of the claims in any way. As those skilled in the art will recognize, the invention can be carried out by using alternative features. Similarly, features introduced by phrases such as “in embodiments of the invention” are intended to be optional features and do not limit any alternative embodiments of the invention, the scope of the invention, or the possibility of combining features introduced in this way with other optional or non-optional features of the invention.

[0155] In summary, and without excluding other possible embodiments, the following embodiments are conceivable:

[0156] Example 1: A method for sorting grain seeds, wherein the grain seeds comprise light-colored grain seeds and / or colorless grain seeds as well as colored seeds, particularly seeds containing a blue aleurone layer, or wherein the grain seeds comprise darker and lighter-colored seeds, particularly seeds containing a dark blue aleurone layer and seeds containing a light blue aleurone layer, the method comprising:

[0157] i. Supplying a seed stream to a sorting station, the sorting station including at least one backlighting device for backlighting the seeds of the seed stream and at least one camera for capturing at least one image of the backlit seeds.

[0158] ii. Use the camera to take at least one image of the backlit seeds of the seed stream;

[0159] iii. Automatically identify the seeds to be sorted from the seed stream from the images captured in step ii; and

[0160] iv. Automatically eject the identified seeds to be sorted from the seed stream.

[0161] The at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space.

[0162] Example 2: According to the method described in the foregoing embodiments, in step iii., the seeds to be sorted from the seed stream are identified by recognizing objects in the image that cumulatively satisfy the following conditions:

[0163] These objects have predefined color coordinates, specifically color coordinates within a predefined range in the HSL color space, and

[0164] These objects have one or more of the following: predefined area, predefined size, predefined diameter, predefined equivalent diameter, and predefined shape.

[0165] Example 3: According to the method described in the foregoing embodiments, the identification of seeds includes determining the color coordinates and area of ​​objects in the image, wherein...

[0166] - If the determined color coordinates are within the predefined range in the HSL color space, then the object is determined to have predefined color coordinates;

[0167] - If the determined area exceeds the area threshold, the object is determined to have a predefined area.

[0168] Example 4: According to the method described in the foregoing embodiments, wherein, in the HSL color space, the predefined range has H coordinates from 0 to 85, specifically from 5 to 65 and / or from 19 to 79, S coordinates from 15 to 85, specifically from 18 to 82 and / or from 19 to 83, L coordinates 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, from 400 to 1500 pixels, or from 500 to 1000 pixels, specifically in the range of 600 to 850 pixels, more specifically in the range of 650 to 700 pixels, and most specifically 700 pixels.

[0169] Example 5: The method according to any one of the preceding two examples, wherein the identification of the seeds to be sorted includes determining at least one identification parameter, the identification parameter including weighting the color coordinates with the area of ​​the object in the image, specifically by using the product of these color coordinates and the area of ​​the object in the image.

[0170] Example 6: The method according to any one of the foregoing embodiments, wherein the at least one backlight has an H coordinate in the HSL color space that is within at least one range selected from the group consisting of:

[0171] -0 ≤ H < 25 or H ≥ 310, specifically excluding the range 332 ≤ H ≤ 338;

[0172] -70 ≤ H ≤ 150;

[0173] -25 ≤ H ≤ 70.

[0174] Example 7: The method according to any one of the preceding embodiments, wherein the at least one backlight has an S coordinate of 0.25 ≤ S ≤ 1.0, specifically 0.5 ≤ S ≤ 1.0, and more specifically 0.75 ≤ S ≤ 1.0 in the HSL color space.

[0175] Example 8: The method according to any one of the preceding examples, wherein the at least one backlight has an S coordinate of 0.25 ≤ S ≤ 1.0 and an H coordinate of 25 ≤ H ≤ 70 in the HSL color space.

[0176] Example 9: According to the method described in the foregoing examples, the at least one backlight has an L coordinate of 0.35 ≤ L ≤ 0.78 in the HSL color space.

[0177] Example 10: The method according to any one of the preceding embodiments, wherein the at least one backlight has an S coordinate of 0.33 ≤ S ≤ 1.0 and an H coordinate of 70 ≤ H ≤ 150 in the HSL color space.

[0178] Example 11: According to the method described in the foregoing embodiments, the at least one backlight has an L coordinate of 0.28 ≤ L ≤ 0.63 in the HSL color space.

[0179] Example 12: The method according to any one of the preceding embodiments, wherein the at least one backlight has an S coordinate of 0.40 ≤ S ≤ 1.0 and an H coordinate of 25 ≤ H or H ≥ 310 in the HSL color space.

[0180] Example 13: According to the method described in the foregoing examples, the at least one backlight has an L coordinate of 0.13 ≤ L ≤ 0.78 in the HSL color space.

[0181] Example 14: The method according to any one of the preceding embodiments, wherein the at least one backlight has an H coordinate of 111, an S coordinate of 1.0 and an L coordinate of 0.48 in the HSL color space.

[0182] Example 15: The method according to any one of the preceding embodiments, wherein the at least one backlight has an H coordinate of 71, an S coordinate of 1.0 and an L coordinate of 0.48 in the HSL color space.

[0183] Example 16: The method according to any one of the preceding embodiments, wherein the at least one backlight has an H coordinate of 340, an S coordinate of 0.75, and an L coordinate of 0.68 in the HSL color space.

[0184] Example 17: The method according to any one of the preceding examples, wherein the at least one backlight has an H coordinate of 47, an S coordinate of 1.0 and an L coordinate of 0.59 in the HSL color space.

[0185] Example 18: The method according to any one of the preceding examples, wherein the seeds to be separated from the seed stream are seeds containing a blue aleurone layer.

[0186] Example 19: The method according to any one of the foregoing 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 image stream of the seed stream is captured, and wherein in step iii., the image stream is continuously evaluated to continuously identify the seeds to be sorted from the seed stream.

[0187] Example 20: The method according to any one of the foregoing embodiments, wherein a batch of seeds is provided, and wherein method steps i.-iv. are repeated on the batch of seeds, specifically at least twice, wherein in each repetition, the batch is reduced by the seeds sprayed in the previous run of step iv.

[0188] Example 21: The method according to any one of the foregoing examples, wherein the grain seeds include, specifically, wheat grains.

[0189] Example 22: The method according to any one of the foregoing embodiments, wherein the method includes at least one backlight identification step, specifically at least one backlight identification step prior to step i, wherein the backlight identification step includes identifying at least one backlight to be used for sorting.

[0190] Example 23: According to the method described in the foregoing examples, the backlight recognition step includes performing multiple sortings using multiple backlights, each sorting including using equal portions of the seed to be sorted.

[0191] Example 24: The method according to the foregoing embodiments, wherein the at least one backlight is determined via at least one quality control step, the quality control step including evaluating sorting purity across multiple backlights, or including evaluating sorting purity and the amount of loss of desired seeds in sorting across multiple backlights.

[0192] Example 25: The method according to any one of the foregoing embodiments, wherein, in step iii., the seeds to be sorted from the seed stream are colored seeds.

[0193] Example 26: The method according to any one of the preceding embodiments, wherein the method includes performing steps i. to iv. using at least one first backlight, the first backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the method further includes repeating steps i. to iv. on one of the sprayed fractions or the retained fractions of seeds using at least one second backlight, the second backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the second backlight is different from the first backlight.

[0194] Example 27: The method according to the foregoing examples, wherein the first backlight is used for the first purpose of sorting grain seeds.

[0195] Example 28: The method according to the foregoing embodiments, wherein the first objective of sorting grain seeds includes at least one objective selected from the group consisting of: maximizing the yield of colorless grain seeds; maximizing the yield of colored grain seeds; maximizing the purity of colorless grain seeds, specifically by removing impurities such as weed seeds, ergot-infected seeds, and / or other plant parts; maximizing the purity of colored grain seeds; reducing the number of colored seeds among colorless seeds; reducing the number of colorless seeds among colored seeds; maximizing the purity of colorless seeds while minimizing colorless seed loss; and minimizing the loss of colorless seeds while minimizing the loss of colorless seeds. Maximizing the purity of colored seeds while minimizing the loss of colored seeds; or any combination thereof; separating dark blue aleurone seeds (with or without colorless seeds) from a seed mixture containing dark blue aleurone seeds and light blue aleurone seeds; separating light blue aleurone seeds (with or without colorless seeds) from a seed mixture containing dark blue aleurone seeds and light blue aleurone seeds; separating dark blue aleurone seeds and light blue aleurone seeds from a seed mixture containing dark blue aleurone seeds, light blue aleurone seeds, and colorless seeds.

[0196] Example 29: The method according to any one of the preceding two examples, wherein the second backlight is used for a second purpose of sorting grain seeds.

[0197] Example 30: The method according to the foregoing embodiments, wherein the second objective of sorting grain seeds includes at least one objective selected from the group consisting of: maximizing the yield of colorless grain seeds; maximizing the yield of colored grain seeds; maximizing the purity of colorless grain seeds; maximizing the purity of colored grain seeds; reducing the number of colored seeds in colorless seeds; reducing the number of colorless seeds in colored seeds; maximizing the purity of colorless seeds while reducing colorless seed loss; maximizing the purity of colored seeds while reducing colored seed loss; or any combination thereof; sorting out dark blue aleurone seeds (with or without colorless seeds) from a seed mixture containing dark blue aleurone seeds and light blue aleurone seeds; sorting out light blue aleurone seeds (with or without colorless seeds) from a seed mixture containing dark blue aleurone seeds and light blue aleurone seeds; sorting out dark blue aleurone seeds and light blue aleurone seeds from a seed mixture containing dark blue aleurone seeds, light blue aleurone seeds and colorless seeds.

[0198] Example 31: The method according to any one of the preceding two examples, wherein the second purpose of sorting the grain seeds is different from the first purpose of sorting the grain seeds.

[0199] Example 32: The method according to any one of the preceding six examples, wherein steps i to iv are repeated using either the ejected fraction or the retained fraction of the seeds, including using a sorting procedure different from the initial sorting in step iii.

[0200] Example 33: The method according to any one of the preceding seven examples, wherein in step iii, the seed to be sorted from the seed stream is identified by recognizing objects in the image with color coordinates having a predefined sorting procedure, specifically objects with color coordinates within a predefined range in the HSL color space.

[0201] Example 34: According to the method described in the foregoing embodiments, the predefined sorting procedure includes at least one procedure selected from the group consisting of the following: the H coordinate includes a range of 354 ≤ H ≤ 45 or a range of H ≥ 354 and H ≤ 54, and the S coordinate is in the range of S... min ≤ S ≤ S max Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The H coordinate includes the range of 354 ≤ H ≤ 45 or the range of H ≥ 345 and H ≤ 79, and the S coordinate is within the range of S... min ≤S ≤ S max Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S maxThe H coordinate includes the range of 354 ≤ H ≤ 45 or the range of H ≥ 345 and H ≤ 45, and the S coordinate is within the range of S... min Within the range of ≤ S ≤ 98, where S min In 55 ≤ S min Within the range of ≤ 62, and the L coordinate is within L min Within the range of ≤ L ≤ 140, where L min In 96≤ L min The range of H coordinates is ≤ 102; the range of H coordinates includes 354 ≤ H ≤ 45 or H ≥ 354 and H ≤ 54; the range of S coordinates is within S... min Within the range of ≤ S ≤ 98, where S min In 55 ≤ S min Within the range of ≤ 62, and the L coordinate is within L min Within the range of ≤ L ≤ 140, where L min In 96 ≤ L min Within the range of ≤ 102; the combination of the first color coordinates and the second color coordinates, wherein the first color coordinates include H coordinates within the range of 354 ≤ H ≤ 45 or 19 ≤ H ≤ 79, S coordinates within the range of 19 ≤ S ≤ 83, and L coordinates within the range of 62 ≤ L ≤ 130, and the second color coordinates include H coordinates within the range of 354 ≤ H ≤ 45 or 5 ≤ H ≤ 65, S coordinates within the range of 18 ≤ S ≤ 82, and L coordinates within the range of 88 ≤ L ≤ 145; H coordinates within the range of 354 ≤ H ≤ 45, S coordinates within the range of 35 ≤ S ≤ 99, and L coordinates within the range of L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min ≤ 30 and where L max It is set to achieve a certain percentage of blowout, such as 78 ≤ L max L ≤ 127 max The H coordinate includes the range of 354 ≤ H ≤ 45, and the S coordinate is within the range of S... min ≤ S ≤ S max Within the range, where S min =0 or 0 ≤ S min ≤ 30, and the L coordinate is in L min ≤ L ≤ L max Within the range, where L min =0 or 0 ≤ L min≤ 30, and where L max and S max It is set to achieve a certain percentage of blowout, such as 78 ≤ L max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The H coordinate includes the range 354 ≤ H ≤ 45, and the L coordinate is within the range L. min ≤ L ≤ L max Within the range, where L min 96 ≤ L min ≤ 110, 96 ≤ Lmin ≤ 102 or 97 ≤ Lmin ≤ 110, such as L of 105 min And L max 138 ≤ Lmax ≤ 145, such as L of 140 max And the S coordinate is in S min ≤S ≤ S max Within the range, where S min 55 ≤ S min ≤ 75, 55 ≤ S min ≤ 62 or 65 ≤ S min ≤ 75, such as 70 for S min And S max 95 ≤ S max ≤ 255, such as 98 or 120 S max Alternatively, the predefined sorting procedure may include at least one procedure with HSL object settings for seeds to be ejected / blown, wherein the H object setting can be any H range as long as it includes 354 ≤ H ≤ 45, such as a sorting procedure with H object settings for any H range for seeds to be ejected / blown but including the range 354 ≤ H ≤ 45, S min and S max The seeds are supplied by a sorting device for the seeds to be blown out, and L min =0 or 0 ≤ L min ≤ 30 and L max Between 85 and 127.

[0202] Example 35: The method according to any one of the preceding two examples, wherein the seeds to be sorted are further identified by recognizing 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.

[0203] Example 36: The method according to any one of the preceding three examples, wherein the seeds to be sorted are further identified by identifying objects in the image whose determined area exceeds an area threshold, the area threshold being 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 pixels, and most specifically 200, 400, 600, 700 or 750 pixels.

[0204] Example 37: The method according to any one of the preceding eleven examples, wherein the method includes repeating steps i to iv using the ejected fraction of the seed and the retained fraction of the seed, respectively.

[0205] Example 38: The method according to any one of the preceding twelve examples, wherein steps i. to iv. are repeated for the retained fraction of the seed using the first backlight, and steps i. to iv. are repeated for the ejected fraction of the seed using the second backlight.

[0206] Example 39: According to the method described in the foregoing embodiments, the first backlight has at least one color in the HSL color space selected from the group consisting of: H coordinate 340, S coordinate 192 and L coordinate 172; H coordinate 71, S coordinate 255 and L coordinate 123.

[0207] Example 40: The method according to any one of the preceding two examples, wherein the second backlight has at least one color in the HSL color space selected from the group consisting of: H coordinate 20, S coordinate 255 and L coordinate 150; H coordinate 111, S coordinate 255 and L coordinate 123; H coordinate 0, S coordinate 250 and L coordinate 110.

[0208] Example 41: The method according to any one of the preceding fifteen examples, wherein steps i. to iv. are repeated for the retained fraction of the seed using the second backlight, wherein steps i. to iv. are repeated for the ejected fraction of the seed using a third backlight, the third backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the third backlight is different from the first backlight and the second backlight.

[0209] Example 42: According to the method described in the foregoing embodiments, the first backlight has an H coordinate of 0, an S coordinate of 255 and an L coordinate of 110 in the HSL color space, or has an H coordinate of 71, an S coordinate of 255 and an L coordinate of 123.

[0210] Example 43: The method according to any one of the preceding two examples, wherein the second backlight has at least one color in the HSL color space selected from the group consisting of: H coordinate 340, S coordinate 192 and L coordinate 172; H coordinate 71, S coordinate 255 and L coordinate 123.

[0211] Example 44: The method according to the foregoing examples, wherein steps i to iv are repeated at least once for the retained fractions of the seed using any second backlight.

[0212] Example 45: The method according to any one of the preceding four examples, wherein the third backlight has at least one color in the HSL color space selected from the group consisting of: H coordinate 20, S coordinate 255 and L coordinate 150; H coordinate 0, S coordinate 255 and L coordinate 110; H coordinate 71, S coordinate 255 and L coordinate 123; H coordinate 111, S coordinate 255 and L coordinate 123.

[0213] Example 46: The method according to any one of the preceding twenty examples, wherein steps i. to iv. are repeated for the retained fraction of the seed using the second backlight, and steps i. to iv. are repeated for the ejected fraction of the seed using the first backlight.

[0214] Example 47: The method according to any one of the preceding twenty-one embodiments further includes repeating steps i. to iv. using retained or ejected fractions of seeds from the ejected fractions of the seeds used in the repeated steps i. to iv., wherein the further repetition uses a fourth backlight having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the fourth backlight is different from the first and / or second backlight.

[0215] Example 48: According to the method described in the foregoing embodiments, the fourth backlight has at least one color in the HSL color space selected from the group consisting of: H coordinate 20, S coordinate 255 and L coordinate 150; H coordinate 0, S coordinate 255 and L coordinate 110; H coordinate 71, S coordinate 255 and L coordinate 123; H coordinate 340, S coordinate 192 and L coordinate 172; H coordinate 111, S coordinate 255 and L coordinate 123.

[0216] Example 49: The method according to any one of the preceding twenty-three examples, wherein a fraction of sprayed and / or retained seeds, comprising colorless grain seeds (specifically composed of such seeds), is added to the repetition of steps i. to iv. using the retained fraction of seeds.

[0217] Example 50: The method according to any one of the preceding twenty-four examples, wherein the retention of the seed is used to repeat steps i. to iv. at least twice.

[0218] Example 51: The method according to any one of the foregoing embodiments, wherein at least step iii., and optionally at least one of steps ii. and iv., is at least one of computer implementation and computer control.

[0219] Example 52: The method according to any one of the preceding embodiments, wherein, in step iii, the seeds to be sorted from the seed stream are identified by a trained artificial neural network (ANN).

[0220] Example 53: The method according to the foregoing embodiments further includes at least one training step, wherein labeled images are used to train the ANN in the training step.

[0221] Example 54: The method according to any one of the preceding two examples, wherein the tagged image includes genotyping data, wherein the genotyping data provides information about seeds with a normal (colorless) aleurone layer and cereal seeds containing a blue aleurone layer, or provides information about cereal seeds containing 1n, 2n, or 3n blue aleurone layers.

[0222] Example 55: The method according to any one of the preceding embodiments, wherein the method further includes at least one second spectral seed sorting step, wherein the second spectral seed sorting step includes determining at least one spectral piece of information about the seeds in the seed stream, such as near-infrared spectral information, wherein the at least one spectral piece of information is used to automatically identify the seeds to be sorted from the seed stream.

[0223] Example 56: According to the method described in the foregoing embodiments, the trained artificial neural network automatically identifies the seeds to be sorted from the seed stream based on the at least one spectral information.

[0224] Example 57: The method according to the foregoing embodiments further includes at least one training step, wherein labeled spectral information is used to train the ANN in the training step.

[0225] Example 58: The method according to any one of the preceding two examples, wherein the tagged spectral information includes genotyping data, wherein the genotyping data provides information about seeds with a normal (colorless) aleurone layer and cereal seeds containing a blue aleurone layer, or provides information about cereal seeds containing 1n, 2n, or 3n blue aleurone layers.

[0226] Example 59: A sorting device for sorting grain seeds, comprising:

[0227] I. At least one seed feeder for supplying a seed stream to at least one sorting station; and

[0228] II. The at least one sorting station includes at least one backlighting device for backlighting the seeds in the seed stream, the sorting station further includes at least one camera for capturing at least one image of the backlit seeds, and the sorting station further includes at least one ejector for ejecting seeds from the seed stream.

[0229] The sorting device is configured to perform the method according to any one of the foregoing embodiments.

[0230] Example 60: The sorting device according to the foregoing embodiments further includes at least one controller, wherein the controller is configured to perform at least step iii of the method, and wherein, optionally, the controller is further configured to control at least one of steps ii and iv of the method.

[0231] Example 61: The sorting device according to any one of the foregoing embodiments involving sorting equipment further includes at least one target sluice and at least one sorting sluice, wherein the seeds ejected in step iv. are collected by the sorting sluice, and wherein the remaining seed stream is collected by the target sluice.

[0232] Example 62: A sorting device according to any one of the foregoing embodiments involving sorting equipment, wherein the sorting station includes at least two cameras configured to capture images of backlit seeds from different angles in space.

[0233] Example 63: The sorting device according to the foregoing embodiments, wherein the sorting station further includes at least two backlight devices, each backlight device being assigned to one of the at least two cameras.

[0234] Example 64: A sorting device according to any one of the preceding two examples, wherein the cameras are configured to capture images of backlit seeds from opposite directions.

[0235] Example 65: A sorting device according to any one of the foregoing embodiments involving sorting equipment, wherein the sorting station is configured such that the seed stream passes through the sorting station between the backlight device and the camera.

[0236] Example 66: A sorting device according to any one of the foregoing embodiments relating to sorting equipment, wherein the injector includes at least one of a mechanical injector having at least one mechanical actuator and a pneumatic injector having at least one nozzle for directing an air jet to seeds to be sorted from a seed stream.

[0237] Example 67: A computer program comprising instructions that, when executed by a sorting device according to any one of the foregoing embodiments relating to a sorting device, particularly by a controller of the sorting device, cause the sorting device to perform at least step iii of the method according to any one of the foregoing embodiments relating to a method, and optionally perform at least one of steps ii and iv of the method.

[0238] Example 68: A computer-readable storage medium, specifically a non-transient computer-readable medium, comprising instructions that, when executed by a sorting device according to any of the foregoing embodiments relating to a sorting device, particularly by a controller of the sorting device, cause the sorting device to perform at least step iii of the method according to any of the foregoing embodiments relating to a method, and optionally perform at least one of steps ii and iv of the method.

[0239] Example 69: The sorting device according to any one of the foregoing embodiments relating to sorting equipment is used for the purpose of selecting from the group consisting of: sorting colored seeds from a seed mixture containing colorless and colored seeds; sorting dark seeds (with or without colorless seeds) from a seed mixture containing dark and lighter-colored seeds; sorting lighter-colored seeds from a seed mixture containing dark and lighter-colored seeds, which may contain colorless seeds; sorting dark and lighter-colored seeds from a seed mixture containing dark, lighter, and colorless seeds; sorting colored seeds from a grain seed mixture containing colorless and colored grain seeds. Grain seeds; sorting colored grain seeds containing a blue aleurone layer from a mixture of colorless grain seeds and grain seeds containing a blue aleurone layer; sorting dark blue aleurone layer seeds (with or without colorless seeds) from a mixture of dark blue and light blue aleurone layer seeds; sorting light blue aleurone layer seeds (with or without colorless seeds) from a mixture of dark blue and light blue aleurone layer seeds; sorting both dark blue and light blue aleurone layer seeds from a mixture of dark blue, light blue, and colorless seeds; using a color space with 70... Using at least one backlight with an H coordinate of ≤ H ≤ 150, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer; using at least one backlight with an H coordinate of 25 ≤ H ≤ 70 in the HSL color space, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer; using at least one backlight with an H coordinate of 0 ≤ H < 25 or H ≥ 310 in the HSL color space, colored grain seeds containing a blue aleurone layer are sorted from a grain seed mixture containing colorless grain seeds and grain seeds containing a blue aleurone layer.

[0240] Example 70: A method according to any one of the foregoing embodiments relating to the method, for separating dark and light-colored seeds from a seed mixture (such as a mixture of blue aleurone seeds and colorless seeds) containing dark and light-colored seeds and colorless seeds, wherein the sorting procedure uses a combination of HSL object settings for dark seeds and HSL object settings for light-colored seeds, such as those provided by the sorting device, for example by sorting dark blue and light blue aleurone seeds from a mixture containing dark blue and light blue aleurone seeds and colorless seeds, by sorting out: (1) objects in an image with H coordinates of 5 ≤ H ≤ 65, S coordinates of 18 ≤ S82, L coordinates of 88 ≤ L ≤ 145 and a size of 200-1500, or 400-1000, or at least 700 pixels, and (2) objects in an image with H coordinates of 19 ≤ H ≤ 79, S coordinates of 19 ≤ S ≤ 83 and L coordinates of 62 ≤ H ≤ 79. The object is L ≤ 130 and has a size of 200-1500, or 400-1000, or at least 700 pixels, or the object in the image has an H coordinate in (1) and (2) that includes the range of 354 ≤ H ≤ 45, preferably using an HSL backlight setting with a hue value of 340, 71, or 0. The method may be performed once or may be repeated multiple times, such as 1-3 times, to maximize the purity of the colorless seed.

[0241] Example 71: The method according to any one of the foregoing embodiments relating to the method, wherein the object setting provided by the sorting device for the colored seeds (such as dark blue aleurone layer seeds or light blue aleurone layer seeds) to be sorted is modified by adjusting the L object setting or by adjusting the S and L object settings to increase the amount of colored seeds sorted, and wherein the H object setting is any H range including 354 ≤ H ≤ 45.

[0242] Example 72: The method according to any one of the foregoing embodiments relating to the method, wherein the amount of dark seeds in batches of dark and light seeds (such as the amount of dark blue seeds containing 3n blue aleurone layers in a batch of blue aleurone layer seeds) is reduced by L of the scheme. min Object settings to reduce, such as L to 0 min or 0 ≤ L min ≤ 60 or 0 ≤ L min ≤ 30, where L max It is set to cause a certain percentage or amount of dark seeds to be blown out, such as using 78 ≤ L max ≤127 L maxThis depends on the percentage or amount of dark seeds (such as 3n blue aleurone layer seeds) to be removed and on the color depth of the seed batch, and the backlight in this method is as described in any of the preceding embodiments, or a white LED backlight or a backlight with an L coordinate of L = 255 (white).

[0243] Example 73: The method according to the foregoing examples, wherein L max Depending on the amount of dark seeds to be removed (such as dark blue seeds containing a 3n blue aleurone layer), and where L is blown out for 10% of dark seeds. max The values ​​are 77 ≤ L ≤ 87, 79 ≤ L ≤ 95 for 15% dark seeds, 81 ≤ L ≤ 100 for 20% dark seeds, 84 ≤ L ≤ 103 for 30% dark seeds, 87 ≤ L ≤ 107 for 40% dark seeds, and 91 ≤ L ≤ 110 for 50% dark seeds.

[0244] Example 74: The method according to any one of the foregoing embodiments relating to the method, wherein the amount of dark seeds (such as dark blue seeds containing 3n blue aleurone layers) in a batch of blue aleurone layer seeds is reduced by decreasing L in the scheme. min and S min Object settings to reduce, such as L to 0 min , or 0 ≤ L min ≤ 60, or 0 ≤ L min ≤ 30, and S of 0 min , or 0 ≤ S min ≤ 40, or 0 ≤ S min ≤ 30, such as using 78 ≤ L max ≤ 120 or 81 ≤ L max ≤ 123 or 78 ≤ L max L ≤ 127 max , and / or 60 ≤ S max ≤ 78 or 60 ≤ S max ≤ 99 of S max This depends on the percentage or amount of dark seeds (such as 3n blue aleurone layer seeds) to be removed and on the color depth of the seed batch, and the backlight in this method is as described in any of the preceding embodiments, or a white LED backlight or a backlight with an L coordinate of L = 255 (white).

[0245] Example 75: The method according to any one of the foregoing embodiments relating to the method, wherein the amount of dark seeds (such as dark blue seeds containing 3n blue aleurone layers) in a batch of blue aleurone layer seeds is reduced by blowing out lighter-colored seeds (such as light blue 1n or 2n blue aleurone layer seeds) using an HSL backlight with the following object setup in the sorting procedure, while retaining the darkest seeds and colorless seeds: 96 ≤ L min ≤ 110 of L min 55 ≤ S min ≤ 75 (e.g., 65 ≤ S) min S ≤ 75) min such as L max For 138 ≤ L max ≤ 145, and S max 95 ≤ S max ≤ 255.

[0246] Example 76: The method according to any one of the foregoing embodiments relating to the method, wherein the sorting method is combined with, or before or after, another sorting method using visual and / or infrared spectral analysis (specifically near-infrared spectral analysis), and / or UV spectral analysis, and / or X-ray spectral analysis, and / or Raman scattering spectral analysis (with or without using trained neural networks to improve sorting efficiency, such as blowing colorless seeds from a mixture of colored and colorless seeds).

[0247] Example 77: According to any one of the foregoing embodiments relating to the method, using the HSL coordinates of the at least one backlight (such as yellow, orange, red and / or green backlight), the purity of the retained colorless seeds (such as when sorting a mixture of blue aleurone layer seeds and colorless seeds) is at least 99%, specifically at least 99.7%, and more specifically, this purity is obtained by repeated blowing, particularly when repeated sorting two or three times, such as said purity can be determined by genotyping the seeds to detect the absence of blue aleurone layer loci.

[0248] Example 78: According to the method or use of the foregoing embodiments, the loss of colorless seeds sprayed during the sorting process to achieve the purity (such as when sorting a mixture of blue aleurone layer seeds and colorless seeds) is less than 35%, less than 30%, less than 25%, less than 20%, less than 15%, less than 10%, or less than 5%.

[0249] Example 79: The method according to any one of the foregoing embodiments relating to the method, wherein the percentage of colorless seeds in the obtained colored seeds (such as blue aleurone layer seeds) is less than 10%, specifically less than 5%, using the HSL coordinates of the at least one backlight (such as yellow, orange, red and / or green backlight), and more specifically, this purity is obtained during repeated sorting, particularly when repeated sorting two to three times.

[0250] Example 80: The method according to any one of the foregoing embodiments of the method, wherein any given HSL coordinate of the at least one backlight in the HSL color space includes any given H coordinate + / - 3 (within the 0-360 H scale), includes any given S coordinate + / - 15 (within the 0-255 S scale), and / or includes any given L coordinate + / - 5 (within the 0-255 L scale).

[0251] Example 81: The method according to any one of the foregoing embodiments relating to the method, wherein the grain seeds comprise light-colored grain seeds and dark-colored grain seeds, or colorless grain seeds and colored seeds. Attached Figure Description

[0252] Further optional features and embodiments will be disclosed in more detail, preferably in conjunction with the dependent claims, in the following description of embodiments. As those skilled in the art will recognize, the respective optional features can be implemented independently and in any feasible combination. The scope of the invention is not limited to the preferred embodiments. Embodiments are schematically depicted in the accompanying drawings. The same reference numerals in these drawings refer to the same or functionally equivalent elements.

[0253] In the attached diagram:

[0254] Figure 1 A side sectional view of an embodiment of a sorting device for sorting grain seeds is shown;

[0255] Figure 2 A flowchart illustrating an embodiment of a method for sorting grain seeds is shown;

[0256] Figure 3A and Figure 3B Exemplary results of a method for sorting grain seeds are shown;

[0257] Figures 4 to 9 Different embodiments of the method for sorting grain seeds are shown; and

[0258] Figure 10 The figure shows a graphical representation of the average distribution of blue seeds in three different grades within the blue grade of the separated spike, based on genotype and color sorting results. Detailed Implementation

[0259] Figure 1 A side sectional view of an exemplary embodiment of a sorting device 110 for sorting grain seeds 112 is shown. The grain seeds 112 comprise colorless grain seeds 114 and seeds 116 containing a blue aleurone layer. For example, the grain seeds 112 may be wheat seeds.

[0260] The sorting equipment 110 includes at least one seed feeder 118 for supplying a seed stream 120 to at least one sorting station 122. For example... Figure 1 As shown, in this exemplary embodiment, the seed feeder 118 may include at least one feed hopper 124 configured to receive a plurality of seeds 112 and controllably supply the seeds 112 to at least one chute 126. The seed feeder 118 may further include at least one vibrating feeder 128 configured to apply vibration to the feed hopper 124, such that the seeds 112 included therein may exit the feed hopper 124 and enter the at least one chute 126.

[0261] Furthermore, the sorting device 110 includes at least one sorting station 122. The sorting station 122 includes at least one backlighting device 130 for backlighting the seeds 112 of the seed stream 120. The sorting station further includes at least one camera 132 for capturing at least one image of the backlit seeds. Figure 1 As can be seen, sorting station 122 can be configured such that seed stream 120 passes through sorting station 122 between backlight device 130 and camera 130. Further, as... Figure 1 As shown, sorting station 122 may include at least one front illumination device 131 for illuminating the seed stream 120 on the front side. The sorting station may include two, four, or even more front illumination devices 131. The front illumination device 131 may include a white LED.

[0262] Furthermore, in Figure 1In an exemplary embodiment, sorting station 122 may include at least two cameras 132. Cameras 132 may be configured to capture images of backlit seeds from different angles in space. Sorting station 122 may further include at least two backlighting devices 130. Each backlighting device 130 may be assigned to one of the at least two cameras 132. Cameras 132 may be configured to capture images of backlit seeds from opposite directions. For example, sorting station 122 may include at least one first backlighting device 134 assigned to at least one first camera 136. The first backlighting device 134 may be arranged on a first side 138 relative to the seed stream 120. The first camera 136 may be arranged on a second side 140 opposite to the first side 138 relative to the seed stream 120. Further, sorting station 122 may include at least one second backlighting device 142 and at least one second camera 144. The second backlighting device 142 may be arranged on the second side 140, and the second camera 144 may be arranged on the first side 138.

[0263] The sorting station 122 further includes at least one injector 146 for ejecting seeds 112 from the seed stream 120. The injector may include a pneumatic injector 148 having at least one nozzle for directing an air jet to seeds to be sorted from the seed stream 120. For example, the injector 146 may include at least one pneumatic injector 148 configured to eject seeds to be sorted from the seed stream 120 by separating the seeds from the seed stream 120 using compressed air directed via the nozzle.

[0264] The sorting device 110 may further include at least one target sluice 150 and at least one sorting sluice 152. In this example, the seeds to be sorted may include seeds 116 containing a blue aleurone layer. Seeds ejected by the ejector 146 may be collected by the sorting sluice 152. The remaining seed stream may be collected by the target sluice 150. When a seed 112 is identified as a seed to be sorted, the sorting station 122, specifically the ejector 146, may be configured to separate the seed 112 from the seed stream into the sorting sluice 152. The sorting device 110 may be configured such that seeds 112 in the seed stream 120 that are not identified as seeds to be sorted may pass through the sorting station 122 toward the target sluice 150.

[0265] The sorting device 110 may further include at least one controller 154. For example... Figure 1 As shown, the controller 154 can be configured to communicate, for example, with other devices of the sorting device 110 (e.g., with the seed feeder 118 and / or with the sorting station 122) via wireless and / or wired means.

[0266] The sorting device 110 is configured to perform the method according to the invention, such as according to Figure 2 Exemplary embodiments and / or methods according to any other embodiments disclosed herein. Therefore, for a description of the method, refer to... Figure 2 The description.

[0267] Figure 2 A flowchart illustrating an embodiment of a method for sorting grain seeds 112 is shown. Grain seeds 112 comprise colorless grain seeds 114 and colored seeds, specifically seeds 116 containing a blue aleurone layer. In this exemplary embodiment, the seeds to be sorted from the seed stream 120 may specifically be seeds 116 containing a blue aleurone layer. The method may further include using a sorting device 110, such as... Figure 1 An exemplary embodiment of the sorting device 110 shown.

[0268] The method includes the following steps, which can be performed in a given order. However, different orders are also possible. In particular, one, more than one, or all of the method steps can be performed once or repeatedly. Further, these method steps can be performed sequentially, or alternatively, one or more method steps can be performed in a time-overlapping manner or even in a parallel and / or combined manner. The method may further include additional method steps not listed.

[0269] The method includes:

[0270] i. (indicated by reference numeral 156) The seed stream 120 is supplied to the sorting station 122, which includes at least one backlighting device 130 for backlighting the seeds 112 of the seed stream 120, and at least one camera 132 for capturing at least one image of the backlit seeds.

[0271] ii. (Indicated by reference numeral 158 in the figure) Take at least one image of the backlit seeds of the seed stream 120 using camera 132;

[0272] iii. (Indicated by reference numeral 160) Automatically identify the seeds to be sorted from seed stream 120 from the images captured in step ii.; and

[0273] iv. (Indicated by reference numeral 162) Automatically eject the identified seeds to be sorted from the seed stream 120.

[0274] The at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space.

[0275] The at least one backlight may specifically have an H coordinate in the HSL color space that is within at least one range selected from a group of the following:

[0276] -0 ≤ H < 25 or H ≥ 310, specifically excluding the range 332 ≤ H ≤ 338;

[0277] -70 ≤ H ≤ 150;

[0278] -25 ≤ H ≤ 70.

[0279] Specifically, H coordinates in the range of 0 ≤ H < 25 or H ≥ 310 may include red backlight. H coordinates in the range of 70 ≤ H ≤ 150 may include green backlight. H coordinates in the range of 25 ≤ H ≤ 70 may include yellow backlight. The at least one backlight may have an S coordinate in the HSL color space of 0.25 ≤ S ≤ 1.0, specifically 0.5 ≤ S ≤ 1.0, and more specifically 0.75 ≤ S ≤ 1.0.

[0280] In step iii., the seeds to be sorted from the seed stream 120 can be identified by recognizing objects in the image that cumulatively meet the following conditions:

[0281] These objects have predefined color coordinates, specifically color coordinates within a predefined range in the HSL color space, and

[0282] These objects have one or more of the following: predefined area, predefined size, predefined diameter, predefined equivalent diameter, and predefined shape.

[0283] Specifically, seed identification can include determining the color coordinates and area of ​​objects in an image, where

[0284] - If the determined color coordinates are within the predefined range in the HSL color space, then the object is determined to have predefined color coordinates;

[0285] - If the determined area exceeds the area threshold, the object is determined to have a predefined area.

[0286] Furthermore, the method can specifically be a continuous method. Specifically, a continuous method for sorting grain seeds 112 may include performing these method steps repeatedly and at least partially overlapping in time. In step i., a continuous stream of seeds may be supplied to sorting station 122. In step ii., a continuous stream of images of seed stream 120 may be captured. In step iii., the image stream may be continuously evaluated to continuously identify seeds to be sorted from seed stream 120.

[0287] For example, in this method, a batch of seeds can be provided. This batch of seeds can be repeatedly subjected to method steps i.-iv., specifically at least twice. In each repetition, the batch may be reduced due to the seeds sprayed in the previous round of step iv. Therefore, by repeatedly performing method steps i.-iv. on this batch of seeds, the purity of the seeds passing through the sorting station can be improved.

[0288] Figure 3A and Figure 3B Exemplary results of a method for sorting grain seeds 112 are shown. Specifically, Figure 3A and Figure 3B An exemplary result of the method for sorting grain seeds 112 obtained in Example 1 is shown, which will be described in further detail below. For a detailed description of this example, refer to the description of Example 1.

[0289] Figure 3A A polar plot of the HSL color space is shown, where the H coordinates are represented as angles of 164° around the center of the polar plot, where H ∈ [0, 360°), and the L coordinates are shown in relative form as distances from the center, where L ∈ [0, 1]. Figure 3A In the image, box 168 highlights the H and S coordinates of the backlight that achieved the highest relative purity in sorting seeds 116 containing a blue aleurone layer, as found in Example 1. Figure 3A It can be seen that a backlight with an H coordinate between H=150 and H=310 is likely suitable, excluding the range of 332 ≤ H ≤ 338. Specifically, the optimal backlight color can include at least one of the following: a yellow backlight with an H coordinate of 47, an S coordinate of 1.0, and an L coordinate of 0.59; a green backlight with an H coordinate of 111, an S coordinate of 1.0, and an L coordinate of 0.48; a green backlight with an H coordinate of 71, an S coordinate of 1.0, and an L coordinate of 0.48; and a red backlight with an H coordinate of 340, an S coordinate of 0.75, and an L coordinate of 0.68. These suitable backlight ranges are within... Figure 3B It is shown in more detail below.

[0290] Specifically, Figure 3B It shows Figure 3A The image highlights the backlight color. Figure 3A The graph shows the L coordinates of the backlight (expressed in absolute and relative values) on the x-axis (170) and the H coordinates of the backlight on the y-axis (172). Additionally, in Figure 3B In the chart, the RBG values ​​corresponding to each HSL value are marked on the y-axis at 172.

[0291] Figures 4 to 9Different embodiments of a method for sorting grain seeds are shown. In this method, a mixed fraction 200 comprising colorless grain seeds 114 and seeds 116 containing a blue aleurone layer (specifically composed thereof) can be sorted. Figures 4 to 9 The illustrated embodiment specifically includes: performing steps i. to iv. using at least one first backlight 202, the first backlight 202 having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space; and further repeating steps i. to iv. using at least one second backlight 208 on one of the ejected seed fraction 204 or the retained seed fraction 206, the second backlight 208 having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the second backlight 208 is different from the first backlight 202. Figures 4 to 9 In the diagram, each box with a cross arrow represents a single round of sorting performed using steps i. to iv. with a specific backlight color, where the thick arrow represents the stream of seeds retained in stage 206, and the thin arrow represents the stream of seeds ejected in stage 208. Figures 4 to 9 In the diagram, the blue aleurone layer seed 116 can be a 3n blue aleurone layer seed (dark blue, bisomatic embryo), as shown in the seed drawing with horizontal stripes; or it can be a 2n blue aleurone layer seed (blue, monomer), as shown in the seed drawing with black dots; or it can be a 1n blue aleurone layer seed (light blue, monomer), as shown in the seed drawing with diagonal stripes (the colorless seed 114 is indicated by the seed drawing without stripes or dots (white surface)).

[0292] exist Figures 4 to 7 In one embodiment, steps i. to iv. can be repeated for the retained fraction 206 of the seed using a first backlight 202, wherein steps i. to iv. can be repeated for the ejected fraction 204 of the seed using a second backlight 208. Figures 4 to 9 In each diagram, the first and further sorting steps in the left triangle represent sorting steps in a stream of colorless (e.g., "white" male-sterile seeds lacking a blue aleurone layer) seeds (the aim is to obtain purer colorless seeds (reducing / removing colored seeds, such as blue aleurone layer seeds)). The first and further sorting steps in the right triangle represent sorting steps in a stream of colored (e.g., fertile seeds with a blue aleurone layer) seeds (the aim is to obtain purer colored seeds (reducing / removing colorless seeds, as well as dark blue 3n blue aleurone layer seeds)). Figures 4 to 9 The numbers 1, 2, 3, and 4 inside the middle circle represent the sorting rounds.

[0293] Figure 4A first embodiment of a method for sorting grain seeds by repeating steps i. to iv. is shown. In this exemplary embodiment, a first backlight 202 having an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172 in the HSL color space is used to perform a first round of sorting 210 on a mixed fraction 200 of grain seeds. For this first round of sorting 210, in step iii., seeds to be sorted from the seed stream can be identified by recognizing objects with color coordinates of a predefined sorting protocol in the image, specifically objects with color coordinates within a predefined range in the HSL color space. For example, a predefined sorting procedure may include a combined procedure with a first color coordinate and a second color coordinate, wherein the first color coordinate includes an H coordinate in the range of 19 ≤ H ≤ 79, an S coordinate in the range of 19 ≤ S ≤ 83, and an L coordinate in the range of 62 ≤ L ≤ 130, and the second color coordinate includes an H coordinate in the range of 5 ≤ H ≤ 65, an S coordinate in the range of 18 ≤ S ≤ 82, and an L coordinate in the range of 88 ≤ L ≤ 145. Figure 4 As shown, the method may include repeating steps i. to iv. at least once, preferably twice, more preferably three times, using the retained seed grade 206 and the first backlighting 202, along with the combined sorting procedure, in a second round of sorting 212 to obtain a high-purity sterile white seed grade with a purity higher than 99%. The ejected seed grade 204 in the second round of sorting 212 may be the discarded seed grade 214 of cereal seeds.

[0294] Furthermore, Figure 4 The embodiment may further include a third round of sorting 216 using the ejected fractions 204 from the seeds of the first round of sorting 210. The third round of sorting 216 may include using a second backlight 208 having an H coordinate of 20, an S coordinate of 255, and an L coordinate of 150 in the HSL color space. Further, for this third round of sorting 216, in step iii., the seeds to be sorted from the seed stream can be identified by recognizing objects in the image with color coordinates having a predefined sorting protocol, specifically objects with color coordinates within a predefined range in the HSL color space. For example, the predefined sorting protocol may include H coordinates in the range of H ≥ 354 and H ≤ 45, and L coordinates having an H coordinate of 20, an S coordinate of 255, and an L coordinate of 150 in the HSL color space. min ≤ L ≤ L max The coordinates of L within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max(The S-coordinates are provided by the sorting equipment for the dark seeds to be sorted); or may include H coordinates in the range of H ≥ 354 and H ≤ 45, in the S... min ≤ S ≤ S max The S coordinates within the range, where S min =0 or 0 ≤ S min ≤ 30, and in L min ≤ L ≤ L max The coordinates of L within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The ejected fraction 204 of the seeds from the third round of sorting 216 can be part of the discard fraction 214 of the grain seeds. The retained fraction 206 of the seeds from the third round of sorting 216 can be used for a further fourth round of sorting 218, wherein, for the fourth round of sorting 218, a second backlight 208 and a further predefined sorting procedure can be used. In the fourth round of sorting 218, the predefined sorting procedure in HSL space can have: H coordinates including the range of H ≥ 354 and H ≤ 45; in L... min ≤ L ≤ L max The coordinates of L within the range, where L min 96 ≤ L min ≤ 110, such as L of 105 min And L max For 138 ≤ L max ≤ 145, such as L of 140 max ; and in S min ≤ S≤ S max The S coordinates within the range, where S min 55 ≤ S min ≤ 75, such as 65 ≤ S min ≤ 75, such as 70 of S min And S max 95 ≤ S max ≤ 255, such as 98 or 120 S maxThe retained fraction 206 of the grain seeds from the fourth round of sorting 218 can be part of the discard fraction 214 of the grain seeds. The ejected fraction 204 of the seeds from the fourth round of sorting 218 can provide a high-purity grain seed retention fraction with a purity of over 95% for seeds 116 containing a blue aleurone layer.

[0295] Figure 5 A second embodiment of a method for sorting grain seeds by repeating steps i. to iv. is shown. Figure 5 The embodiments largely correspond to Figure 4 The embodiments are described below. Therefore, for a detailed description thereof, please refer to the following: Figure 4 The description is as follows. However, in this exemplary embodiment, the first backlight 202 in the first round sorting 210 and the second round sorting 212 may have an H coordinate of 71, an S coordinate of 255, and an L coordinate of 123 in the HSL color space.

[0296] Figure 6 and Figure 7 The third and fourth embodiments of the method for sorting grain seeds by repeating steps i. to iv. are shown. Figure 6 and Figure 7 The embodiments respectively largely correspond to Figure 4 and Figure 5 The embodiments are described below. Therefore, for a detailed description thereof, please refer to the following: Figure 4 and Figure 5 Description: In Figure 6 In the third embodiment, the first backlight 202 in the first round of sorting 210 and the second round of sorting 212 can have the same color space as in the HSL color space. Figure 4 The embodiment has a similar H coordinate of 340, an S coordinate of 192, and an L coordinate of 172. In Figure 7 In the fourth embodiment, the first backlight 202 in the first round of sorting 210 and the second round of sorting 212 can have the same color space as in the HSL color space. Figure 5 The embodiments are similar in having H coordinates of 71, S coordinates of 255, and L coordinates of 123. In both embodiments, the second backlight 208 used in the third round of sorting 216, which uses the ejected fractions 204 from the seeds of the first round of sorting 210, can have H coordinates of 20, S coordinates of 255, and L coordinates of 150 in the HSL color space.

[0297] However, in Figure 6 and Figure 7In some embodiments, the method may include repeating steps i. to iv. using retained fractions 206 from the seeds that were sprayed using seed fractions 204, repeating the steps i. to iv., wherein this further repetition may use a fourth backlight 220 having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The fourth backlight 220 may specifically differ from the first backlight 202 and / or the second backlight 208. In these exemplary embodiments, the fourth backlight 220 may be used for a fourth round of sorting 218. The fourth backlight 220 may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110 in the HSL color space, or an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172. The predefined sorting procedure in this fourth round of sorting 218 may, for example, include the aforementioned grouping procedure.

[0298] Furthermore, such as Figure 6 and Figure 7 As shown, the retained seed fraction 206, comprising colorless grain seeds 114 (specifically composed of them), can be added to a repetition of steps i. to iv. using the retained seed fraction 206. Specifically, in these examples, the retained seed fraction 206 of the fourth round of sorting 218, comprising colorless grain seeds 114 (specifically composed of them), can be added to a repetition of steps i. to iv. using the retained seed fraction 206 of the first round of sorting 210. In other words, the retained seed fraction 206 of the fourth round of sorting 218 can be added to the second round of sorting 212.

[0299] Figure 8 A fifth embodiment of a method for sorting grain seeds by repeating steps i. to iv. is shown. In this exemplary embodiment, steps i. to iv. can be repeated for the retained fraction 206 of seeds using a second backlight 208, wherein steps i. to iv. can be repeated for the ejected fraction 204 of seeds using a third backlight 222, the third backlight 222 having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space. The third backlight 202 may specifically differ from the first backlight 202 and the second backlight 208.

[0300] For example, the first backlight 202 used in the first round of sorting 210 may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110 in the HSL color space. The predefined sorting procedure used in step iii. of the first round of sorting 210 may include, for example, the aforementioned grouping procedure. The second round of sorting 212 may be performed using the retained grade 206 from the seed of the first round of sorting 210. The second backlight 208 used in the second round of sorting 212 may have an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172 in the HSL color space. The same grouping procedure may be used as the predefined sorting procedure in step iii. of the second round of sorting 212. The second backlight 208 may be used to repeat steps i. to iv. at least once, preferably at least once, more preferably twice, and even more preferably three times, using the retained grade 208 from the seed of the first round of sorting 210. The ejected fraction 204 of seeds in the second round of sorting 212 can be part of the discard fraction 214 of grain seeds.

[0301] Furthermore, such as Figure 8 As shown, a third round of sorting 216 can be performed using the ejected fractions 204 from the seeds of the first round of sorting 210. In this third round of sorting 216, the third backlight 222 can have an H coordinate of 20, an S coordinate of 255, and an L coordinate of 150 in the HSL color space. Further, for this third round of sorting 216, in step iii., the predefined sorting procedure can include H coordinates in the range of H ≥ 354 and H ≤ 45, and L coordinates in the range of S ≤ 45. min ≤ L ≤ L max The coordinates of L within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max (The S-coordinates are provided by the sorting equipment for the seeds to be sorted); or may include H coordinates in the range of H ≥ 354 and H ≤ 45, in the S... min ≤ S ≤ S max The S coordinates within the range, where S min =0 or 0 ≤ S min ≤ 30, and in L min ≤ L ≤ L max The coordinates of L within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S maxIt is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The ejected fraction 204 of the seeds from the third round of sorting 216 can be part of the discard fraction 214 of the grain seeds. The retained fraction 206 of the seeds from the third round of sorting 216 can be used for a further fourth round of sorting 218. A fourth backlight 220 can be used for the fourth round of sorting 218; for example, an H coordinate of 71, an S coordinate of 255, and an L coordinate of 123. In the fourth round of sorting 218, the predefined sorting procedure in the HSL space can have: H coordinates in the range of H ≥ 345 and H ≤ 45; and L coordinates in the range of H ≥ 345 and H ≤ 45. min ≤ L ≤ L max The coordinates of L within the range, where L min 96 ≤ L min ≤ 110, such as L of 105 min And L max For 138 ≤ L max ≤ 145, such as L of 140 max ; and in S min ≤ S ≤ S max The S coordinates within the range, where S min 55 ≤ S min ≤ 75, such as 65 ≤ S min ≤ 75, such as 70 of S min And S max 95 ≤ S max ≤ 255, such as 98 or 120 S max .

[0302] like Figure 8 As shown, this exemplary embodiment can also be used to obtain high-purity sterile white seed fractions with a purity higher than 99% via a second round of sorting 212, and to obtain high-purity cereal seed maintainer fractions with a purity higher than 95% containing a blue aleurone layer via a fourth round of sorting 218.

[0303] Figure 9 A sixth embodiment of a method for sorting grain seeds by repeating steps i. to iv. is shown. In this exemplary embodiment, steps i. to iv. can be repeated for the retained fraction 206 of seeds using a second backlight 208, and further, steps i. to iv. can also be repeated for the ejected fraction 204 of seeds using a first backlight 202.

[0304] For example, the first backlight 202 used in the first round of sorting 210 may have an H coordinate of 0, an S coordinate of 255, and an L coordinate of 110 in the HSL color space. The predefined sorting procedure used in step iii. of the first round of sorting 210 may include, for example, the aforementioned grouping procedure. The second round of sorting 212 may be performed using the retained grade 206 from the seed of the first round of sorting 210. The second backlight 208 used in the second round of sorting 212 may have an H coordinate of 340, an S coordinate of 192, and an L coordinate of 172 in the HSL color space. The same grouping procedure may be used as the predefined sorting procedure in step iii. of the second round of sorting 212. The second backlight 208 may be used to repeat steps i. to iv. at least once, preferably at least once, more preferably twice, and even more preferably three times, using the retained grade 208 from the seed of the first round of sorting 210. The ejected fraction 204 of seeds in the second round of sorting 212 can be part of the discard fraction 214 of grain seeds.

[0305] Furthermore, such as Figure 9 As shown, the first backlight 202 and the combined sorting procedure can also be used to repeat steps i. to iv. in the third round of sorting 216 for the ejected fraction 204 of seeds from the first round of sorting 210. The ejected fraction 204 of seeds from the third round of sorting 216 can be used for a further fourth round of sorting 218, while the retained fraction 206 of seeds from the third round of sorting 216, which includes colorless grain seeds 114 (specifically composed of them), can be added to the second round of sorting 212. Further, in the fourth round of sorting 218, a fourth backlight 220 can be used, which has an H coordinate of 20, an S coordinate of 255, and an L coordinate of 150 in the HSL color space. For this fourth round of sorting 218, in step iii., the predefined sorting procedure can include H coordinates in the range of H ≥ 354 and H ≤ 45, and L coordinates in the range of H ≥ 354 and S coordinates ≤ 45. min ≤ L ≤ L max The coordinates of L within the range, where L min =0 or 0≤ L min ≤ 30, and where L max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max (The S-coordinates are provided by the sorting equipment for the seeds to be sorted); or may include H coordinates in the range of H ≥ 354 and H ≤ 45, in the S... min ≤ S ≤ S max The S coordinates within the range, where S min =0 or 0 ≤ S min ≤ 30, and in L min ≤ L ≤ Lmax The coordinates of L within the range, where L min =0 or 0 ≤ L min ≤ 30, and where L max and S max It is set to allow a certain percentage of dark seeds to be blown out, such as 78 ≤ L. max L ≤ 127 max and 60 ≤ S max ≤ 99 of S max The ejected fraction 206 from the fourth round of sorting 218 can be part of the discard fraction 214 of the grain seeds.

[0306] like Figure 9 As shown, this exemplary embodiment can also be used to obtain high-purity sterile white seed fractions with a purity higher than 99% via a second round of sorting 212, and to obtain high-purity cereal seed maintainer fractions with a purity higher than 95% containing a blue aleurone layer via a fourth round of sorting 218.

[0307] In addition, Figures 4 to 9 In the exemplary embodiment shown, any remaining colorless seeds (lacking a blue aleurone layer) may be blown out before, during, or after the sorting steps performed in the seed sorting stream of the breeding line, using any other sorting method, such as visual and / or infrared spectroscopy (specifically near-infrared spectroscopy), and / or UV spectroscopy, and / or X-ray spectroscopy, and / or Raman scattering spectroscopy, and / or multispectral analysis, with or without the use of trained neural networks to improve the blowing out of colorless seeds. These other sorting methods may be performed on sorting equipment that is the same as or different from the method of the present invention (e.g., a Cimbria SEA.IQ PLUS or an H-series sorter from AnySort, with or without adjustable backlight color).

[0308] Figure 10 A graphical representation of the average distribution of blue seeds in three different grades within the blue grade of the segregating spike, based on genotype and color sorting results. The bars below indicate a brightness scale from light (255) to dark (0), and L... max Approximate positions for 89, 87, and 85 (the marked 5%, 10%, and 20% are blown-out targets expressed as percentages by weight). L: Bright, D: Dark, X-axis: Relative brightness (RL), Y-axis: Relative amount (RA) of different seed types with 1n blue aleurone layers, 2n blue aleurone layers, and 3n blue aleurone layers.

[0309] The invention is further illustrated by the following examples.

[0310] Example 1

[0311] In the first instance, the method was performed using the same procedure to sort objects in the image with (1) H coordinates ranging from 5 to 65, S coordinates ranging from 18 to 82, and L coordinates ranging from 88 to 145 (for light blue seeds) and dimensions greater than 700 pixels; and objects in the image with (2) H coordinates ranging from 19 to 79, S coordinates ranging from 19 to 83, and L coordinates ranging from 62 to 130 (for dark blue seeds) and dimensions greater than 700 pixels. These two object specifications (1) and (2) were combined to blow out both light blue (LB) seeds (1) and dark blue (DB) seeds (2) (hence, this is referred to as the LB_DB or DB_LB blowing method). The method was performed once using an ASM® EUREKA sorter (only one blowing operation) to test the effect of different colored backlights on the purification of white seeds in a mixed batch of colored seeds. The percentage (represented by "P" in the table, Impurity Percentage) refers to the contamination of blue seeds in the white fraction detected by a multispectral imaging device as a quality control step. The multispectral imaging device in this example is equipped with an automatic feeder and a blue conveyor belt. The blue of the conveyor belt as a background can hinder accurate differentiation between white and light blue seeds. Therefore, in this analysis, some light blue seeds may be classified as white, leading to an overestimation of purity. Therefore, the purity specified in Tables 1 and 2 should be understood as relative purity, having the same systematic bias in all test backgrounds, thus allowing them to be compared relative to each other. Control analyses using a multispectral imaging device can also be performed using different colors of the conveyor belt. A black conveyor belt was found to provide the most accurate results for visual inspection. Lower percentages (below P) indicate higher sorting purity. The amount of "white" seeds (represented by W in the table) is measured in grams (g). In this example, the colorless "white" seeds have red seed coats. Optimal backlighting can be identified by high yield and high purity (i.e., a low percentage of blue seeds), preferably above 99%, more preferably above 99.7%, which can be specifically achieved by repeating the method at least 2 or 3 times. The method is performed using approximately 200g of a mixture of wheat seeds, comprising colorless cereal seeds and seeds containing a blue aleurone layer. Quality control of approximately 600 seeds is performed using a multispectral imaging device (Videmeter®).

[0312] Table 1 Experimental results of Example 1

[0313]

[0314] Table 1 Experimental results of Example 1 (continued)

[0315]

[0316] These results are summarized in Figure 3A and Figure 3B As can be seen from the attached figures and the experimental data above, colored backlights such as magenta (hue 300), blue (hue 240), and cyan (hue 180) are unsuitable for achieving pure white seed fractions compared to colored backlights such as yellow (e.g., hue 47), green (e.g., hue 71), and red (e.g., hue 340), as indicated by the significantly higher percentage of remaining blue seeds in the white fraction after one round of sorting. Generally, background colors with hue values ​​between 150 and 310 are unsuitable. In the five different seed batches tested (after one blow-out or sorting run, as used here to test various backlight colors), green, yellow, and red backlights resulted in an average of approximately 2-5% remaining blue seed impurities in the white fraction, depending on the brightness, while other colors showed 6-25% remaining impurities. Each different seed batch used corresponded to a mixture of blue and colorless seeds from different genotypes, where all blue seeds contained the same blue aleurone layer locus.

[0317] Example 2:

[0318] Example 2 essentially corresponds to Example 1. Example 2 was performed to further test the ability of the backlight color from Experiment 1 to promote high-precision sorting, resulting in a relative purity of over 99.7% for the white seed fraction (after at least two sorting runs, compared to other backlight colors), as currently required for certified seed registration. Additionally, the best backlight color in each color group was compared to previously used backlight colors to assess any potential superiority in purity and / or product yield. Using the same procedures as in Example 1, an ASM® EUREKA seed sorter was used to sort genotypes from six different varieties containing both white and blue seeds, with a blue aleurone layer color separation. This sorting method was repeated at least twice for the white seed fraction, and the resulting white fraction was weighed to obtain product yield and its purity analyzed using a multispectral imaging device (Videometer®). At least 600 seeds from each white fraction were analyzed, with impurities expressed as the percentage of blue seeds detected within the white seeds, and the purity of approximately 600 seeds from the resulting white fraction was assessed using a multispectral imaging device, as described in Example 1.

[0319] Table 2 Experimental results of Example 2

[0320]

[0321] Table 2 Experimental results of Example 2 (continued)

[0322]

[0323] Table 3 Experimental results in Example 2 with white and cyan backgrounds and no background lighting ("black").

[0324]

[0325] The results showed that for backlighting with H coordinates of H ≤ 150 or H ≥ 310 (except for H332-338) and L coordinates of 0.2 ≤ L ≤ 0.80 (see Figures 3a and 3b), these genotypes could be sorted to a purity of 99.7% or higher. Genotype 23NGTA004793 was consistently difficult to sort, and the required purity could only be achieved under green backlighting with three consecutive sorting operations. When the purity of all tested genotypes was averaged, yellow backlighting performed best, achieving an average relative purity of 99.86% for white seed grades. It should be noted that not all seed batches performed alike. This may be due to the fact that there are significant differences in color appearance between blue and white seed grades among some of the different genotypes used. For example, genotype 21NGTA004789 appears much darker than, for example, genotype 21NGTA004785. These different genotypes originate from different field-grown varieties and have been infiltrated into these varieties through repeated backcrossing, introducing the same blue aleurone layer gene locus. Therefore, different genotypes also differ in size (thousand-grain weight), shape, or degree of wrinkling. Consequently, the optimal background light may vary depending on the genotype.

[0326] Furthermore, depending on the priorities of seed sorting, such as the priority of purity requirements and / or the quantity of seeds obtained (product yield), a single seed batch may be best suited for sorting under a specific background to produce the highest product yield and a lower purity, such as 99%. Such an example could be the yellow (H, S, L) = (20, 255, 150) genotype 21NGTA004791, in which significantly higher product yields were obtained through two rounds of sorting, and white seed yields exceeded 100 grams, corresponding to a relative purity greater than 99%. Therefore, for important single genotypes, especially in larger batches such as several kilograms or tons, it is recommended to try several established backgrounds on smaller aliquots to identify the optimal background lighting for larger-scale sorting. In another example, such as in a breeding background where hundreds of different small seed batches may need to be sorted, but not all available seeds are required, a background light that allows for consistent high-purity sorting but lower product yields across all different seed batches might be preferred.

[0327] Compared to the suitable backlight colors listed in Table 2, applying white or cyan backlighting (cyan is the recommended default color for the sorter and for sorting these types of seeds) failed to purify the white fraction of both selected seed batches to a sufficient purity. Furthermore, the amount of white seeds obtained after two blowouts was very low (see Table 3). Due to the black paint, the camera background appears black if the backlight is turned off. This background yielded different results depending on the genotype being sorted. In the case of genotype 23NGTA004787, purity was high and product yield was relatively low compared to the suitable background colors listed in Table 2. In the case of other genotypes, purity was low and product yield was very low.

[0328] Example 3:

[0329] Example 3 essentially corresponds to Example 1. In this example, the method for sorting wheat seeds was repeated for different genotypes from which a larger quantity of seeds was available, with three blowouts. Purity analysis was performed by visual inspection of a total of four different images using a multispectral imaging device (Videometer®), with approximately 170–200 seeds from the white fraction captured each time, placed in petri dishes and imaged against a black background. The table lists the amount of blue seeds detected in the total of approximately 700–800 seeds in the white fraction. The high level of accuracy of this image-based purity assessment method was confirmed by cross-validation of genotype purity assessment using SNP markers targeting the BA locus. From the available seed batches, two of the visually lightest and two of the darkest seed batches were selected for blue-white sorting to cover the entire range of seed color variation.

[0330] In Table 4, the number of white seeds is represented by #w, the number of blue seeds by #b, the amount of white seeds (in grams) by A, the average number of blue seeds by ∅#b, and the percentage of impurities by P.

[0331] Table 4 Experimental results of Example 3

[0332]

[0333] Example 4: Using the brightness (low L) method to reduce the blue aleurone layer seed by 3n

[0334] The principle of the brightness method:

[0335] To blow out the darkest seeds from the seed batch, the Lmin object setting is set to 0 (maximum color depth), and the Lmax value is selected based on the percentage of dark blue seeds to be blown out (see [link]). Figure 10 (A virtual example in the text). Because the 3n and 2n Bla levels overlap in color depth (as shown in the example). Figure 10As shown, when more dark blue seeds are blown out, the likelihood of blowing out single, 2n (and lower percentages of 1n) Bla seeds increases. At the same (low) Lmax value, more seeds are blown out from dark seed batches compared to light seed batches, therefore seed batches with significantly different color expressions require different Lmax values. max Only by setting it up can you blow out the same percentage of the deepest blue seed.

[0336] The brightness (or low L) method uses only the brightness aspect of the blue seeds (to be blown out) to differentially target the deepest blue seeds to blow out 3n blue aleurone layer seeds (also referred to herein as double blue (or DB) or diploid seeds (3n refers to the aleurone layer, and diploid / double refers to the embryo)). To achieve this, the L of the procedure is... min Set the object value to 0 and select L to cause the darkest seed to be blown out in the required proportion by weight. max Value. This was tested on seed batches from collected and separated spikelets, with an expected percentage of 10% for double-blue seeds (pre-sorted using the method described above to obtain the blue seed grade). In the ASM® EUREKA color sorter, the backlight HSL settings were set to H20 (0-360 scale) / S255 (0-255 scale) / L150 (0-255 scale), and the blown object settings were: H: 354-54, S... min 35, S max :99,L min :0,L max :x (depending on the percentage to be blown out; for the same percentage of blown out, lighter seed batches require a higher Lmax brightness value compared to darker seed batches). The results are shown in Table 5 below, illustrating the Bla locus copy number analysis of blue seeds from segregating spikes of four different BLA lines and their corresponding DB blown-out fractions (5%, 10%, and 20% by weight) using the brightness method.

[0337] Table 5

[0338]

[0339] In this example, the percentage of double blue seeds in the blue / “white” seeds from the separated spike varied between 12% and 27% (average 17%). The efficiency of the brightness method was at most about 50% (at a blow-out of 5% by weight of blue seeds, the 3n blue aleurone layer in the blow-out fraction was about 50%).

[0340] For the blue fraction from seed batches derived from isolated spikes, it is recommended to blow out approximately 20% by weight to reduce the number of double-blue seeds by about 50%, accompanied by a loss of approximately 15% of the single-blue seeds (1n / 2n blue aleurone layer). However, if the aim is to maximize the propagation rate and efficiency of the maintainer line (reducing the loss of 1n / 2n maintainer line seeds), it is recommended to blow out only 5% to 10% of the blue seeds from isolated spikes. It appears that for blue seed batches derived from isolated spikes, a 5-10% blowout (by weight) can contain more than 50% of the double-blue (3n) seeds.

[0341] In a further example using the above brightness method, it can be seen that L min A value of 0 is best used as a standard for any new seed batch (it works well across different seed batches), L min It can also be improved—for example, for dark seed batches, L min The (object) value can be increased to 30, and to 60 for light-colored seed batches, without significantly affecting the percentage of dark blue seeds blown out. Furthermore, in this (low L) method, the optimal L for blowing out 10% dark blue seeds is... max For 77 ≤ L ≤ 87, for 15% of the seeds blown out, it is 79 ≤ L ≤ 95; for 20% of the seeds blown out, it is 81 ≤ L ≤ 100; for 30% of the seeds blown out, it is 84 ≤ L ≤ 103; for 40% of the seeds blown out, it is 87 ≤ L ≤ 107; and for 50% of the seeds blown out, it is 91 ≤ L ≤ 110 (0-255 L scale).

[0342] Furthermore, when testing other H object settings in the low L method, it was found that virtually any hue object setting, including those ranging from 354 ≤ H to H ≤ 45, also worked well, and even minor variations of + / - 3 applied.

[0343] Example 5: Improving the blowout of 3n blue aleurone layers by using brightness and saturation

[0344] Surprisingly, the results show that simultaneously reducing the saturation object setting in the aforementioned brightness method improves the visual impression of blowing out dark seeds. This fraction generally appears more uniform and deeper. The average values ​​from the video spectrometer support this visual impression. Therefore, the procedure based on both low saturation (S) and low brightness (L) object settings is particularly suitable for blowing out 3n Bla seeds from blue seed fractions.

[0345] In the first procedure for setting up the backlight HSL using H20 / S255 / L150, S min and S minAll object settings are set to 0, and the effect depends on the appearance of the seed batch (for the same percentage of dark blue blown out, a lighter seed batch requires a higher S value compared to a darker seed batch). max and L max (value) and the amount of seeds intended to be blown out from the blue grade, S max Set between 60 and 78, and L max The settings were between 78 and 120 (0-255 scale), and the H object was set to 354-54. The results show that when S... max Value rather than L max When the value is used to limit the amount of seeds blown out under a specific setting, the blown-out seeds appear the darkest, making it the preferred method for blowing out the required amount of dark seeds to minimize the consumption of double blue (double body) seeds.

[0346] When this brightness and saturation (or low L+S) method is compared with the brightness method of Example 4, on the same seed batches of different wheat varieties (containing a mixture of colorless and 1n, 2n and 3n blue aleurone layer seeds, 5 varieties produced in Germany and 2 in France), the L+S method yields a deeper ejected fraction (as measured by video spectrometer), indicating that the L+S method blows out a higher proportion of 3n Bla seeds compared to the brightness method.

[0347] In the second test, seed batches from the collected segregated spikelets were sorted once to separate blue aleurone layer seeds and colorless (lacking a blue aleurone layer, "white") seeds. The blue fraction was equally split, and approximately 10% of the deepest blue seeds were blown out (blown out (g)) using either the brightness method of Example 4 or the L+S method. The percentage of 3n Bla seeds in the deep blue fraction was determined by genotyping ("DB percentage" in Table 6 below). The percentage attenuation rate of 3n Bla seeds was calculated based on the weight and percentage of 3n Bla seeds in the undiminished fraction and the deep blue fraction.

[0348] Consistent with the results from the video spectrometer, the L+S method is more efficient than methods based solely on brightness (low L) for seed batches where the amount of 3n Bla seeds is not limited. In these seed batches (here, lines B and C), within the same / similar blown seed amount (target 10%), the proportion of 3n Bla seeds identified by the L+S method is higher than that of 1n and 2n Bla seeds compared to the brightness method (see Table 6: DB percentage in blown seeds). Therefore, a higher degree of 3n Bla (DB) seed attrition was achieved here using the L+S method.

[0349] Table 6

[0350]

[0351] Different backlight HSL settings were tested in the L+S method. For the light blue seed batch (23NGTA013748), the optimal HSL backlight settings in this method were found to be H20 / S255 / L150, H0 / S255 / L110, H111 / S255 / L123, H71 / S255 / L123 and H47 / S255 / L150.

[0352] In addition, the low-L+S and low-L methods were tested on other wheat lines producing blue seeds with 1n, 2n, and 3n blue aleurone layers, as well as “white” seeds without a blue aleurone layer. These seed batches were derived from wheat lines that had undergone two rounds of continuous breeding (without removal of DB seeds), thus containing a higher percentage of DB seeds. The copy number of the BLA locus was analyzed by PCR in 135–169 seed samples randomly selected from different 15% or 30% deep blue blow-out fractions. Seeds from the reference fraction were analyzed to determine the composition of the blue seed fraction prior to blow-out. Seeds genotyped as 0n were “white” seeds without the BLA locus, seeds genotyped as 1n or 2n were monosomy (the desired maintainer line), and seeds genotyped as 3n were disomy (double blue, 3n blue aleurone layer) seeds (undesirable).

[0353] The results are shown in Table 7a below.

[0354] Table 7a

[0355]

[0356] Highly efficient reduction of DB seeds was achieved for strains ii and iii, with 3n Bla seeds accounting for over 90% (strain ii) and up to 85% (strain iii) of all blown seeds out at 15% by weight. The DB blowing efficiency for strain iii decreased slightly to 76% at 30% blowing. These results imply that, using a low L or low L+S method and blowing out 30% by weight of the darkest seeds, the corresponding proportion of all initially present dimorph / 3n blue seeds removed from the blue seed fraction in a single blow is 45% for strain ii (29% percentage loss of retainer seeds (1n / 2n) during blowing), 61% for strain ii (4% percentage loss of retainer seeds (1n / 2n) during blowing), and 58% for strain iii (12% percentage loss of retainer seeds (1n / 2n) during blowing). The percentage of DB seeds removed by DB blowing is calculated as follows: (blown weight (g) × percentage of DB seeds detected in the blowing) / (amount of blue seeds used for sorting (g) × percentage of DB seeds detected therein). Since no significant decrease in DB depletion efficiency was observed between 15% and 30% of dark blue seeds blown out, it appears that the amount of dark blue seeds blown out could be further increased (e.g., 40% or 50% by weight) to further increase the depletion of binary (3n) seeds.

[0357] Furthermore, testing other H-object settings in the low L+S method revealed that they essentially range from 354 ≤ H to H Any hue setting within the ≤45 range works well, and even minor variations of + / - 3 apply.

[0358] The low-L+S and low-L methods were further tested on the same two lines as described above (lines ii and iii in Table 7a) and on one additional wheat line that produced blue seeds with 1n, 2n, and 3n blue aleurone layers, as well as “white” seeds without a blue aleurone layer. All these seed batches were derived from wheat lines that had undergone two rounds of continuous breeding and therefore contained a higher percentage of DB seeds. To further test the efficacy of the two methods, 40% and 50% of the deepest blue seeds were blown out from lines ii and iii, respectively. In the case of the other line, 20% and 40% of the deepest blue seeds were blown out. The copy number of the BLA locus was analyzed by PCR in samples of 143–164 seeds randomly selected from different deep blue blown-out fractions. Seeds from the reference fraction were analyzed to determine the composition of the blue seed fraction prior to blown out. Seeds genotyped as 0n are "white" seeds without the BLA locus; seeds genotyped as 1n or 2n are monotypic seeds (required maintainer lines); and seeds genotyped as 3n are dimorphic (double blue, 3n blue aleurone layer) seeds (unnecessary). The results are shown in Table 7b below.

[0359] Table 7b

[0360]

[0361] For strains ii and iii, DB reduction efficiency was improved compared to 30% blowout. For strain ii, the L+S method improved from 61% (30% blowout) to 72% at 50% blowout, and to 66% at the brightness method; while for strain iii, the L+S method improved from 58% to 86% at 50% blowout, and to 71% at the brightness method. For new seed batches (strain iv), DB removal was most effective when 40% of the darkest blue seeds were blown out, achieving DB reductions of 79% and 84% of the total seed volume, respectively, regardless of whether the L+S or brightness method was used. Table 7b also shows the percentage of retainer seeds (1n / 2n seeds) lost during blowout (by weight), indicating that the L+S method generally has lower retainer seed loss than the brightness method.

[0362] Example 6: Light Blue (LB) Blowing Method

[0363] The maintainer line fraction in a two-line wheat hybrid system based on the blue aleurone layer should meet two criteria: a) a small amount of dark blue 3nBla (disomy) seeds, and b) a small amount of male-sterile "white" seeds. After the first blue aleurone layer seed blowout, approximately 5–15% of white seeds are typically retained in the blue seed fraction (e.g., using the backlit color of H71 / S255 / L123). The goal of the maintainer line seed stream is to contain less than 5% of "white" male-sterile seeds, thus often requiring a sorting step to remove them. A novel concept of light blue blowout was tested, designed to reduce not only (further) the amount of 3nBla (disomy) seeds but also the amount of white seeds.

[0364] Based on results / knowledge obtained from the low L+S deep blue blowing method, the L and S settings were modified. An attempt was made to obtain a large number of white seeds in the blowing process, as well as seeds that are only optically extremely deep blue in this blowing process, without significant loss of light blue seeds.

[0365] In the first procedure, S max and L max The object settings were fixed at 98 and 140 respectively. For the extreme seed batches tested (shallowest and deepest), S min and L min The values ​​are very similar. For the darkest seed batch, the setting is S. min It is 55 and L min The value is 96; while for the lightest seed batch, the setting is S. min It is 62 and L min The value is 102 (H object settings: 345-45 or 354-54). Typically, with one blow using this method, the percentage of "white" seeds in the blue fraction decreases from 6-15% to 2-3%. The optimal HSL background settings for this method are: H47 / S255 / L150 for H:354-54 objects and H0 / S255 / L110 for H:345-45 objects.

[0366] In one test, different sorting strategies were used to sort pre-sorted blue seeds (blown from a mixture of blue and white, thus rich in blue seeds) from the same wheat strain grown in Gatesleben, Germany and Milly, France. Sorting strategy A for the Milly seeds was the LB-DB blowing method of Example 1, executed 4 times, while sorting strategy B for the Gatesleben seeds was a light blue blowing method, executed 2 times, followed by one execution of the LB-DB method of Example 1. The results are shown in Table 8 below. Both strategies were performed on an ASM® EUREKA sorter using the H71 / S255 / L123 backlight setting, with the HSL object settings for the LB blowing procedure as follows: H: 348-48, S... min :55,S max 140, L min :48,L max :110 (pixel size 1000), and the HSL object settings for the LB-DB blowout method are as described in Example 1.

[0367] Table 8

[0368]

[0369] Therefore, this sorting strategy, including the light blue blowing method, significantly improved the reduction of 3n dark blue seeds (from an average of 38% to 19%).

[0370] In another test, using pre-sorted blue fractions removed from the initial sorting to improve the purity of white seeds, the range of HSL object settings that could be used in the LB blow-out method for different types of blue aleurone layer seed genotypes was tested to blow out light blue seeds while retaining dark blue and white seeds. The procedure was established such that dark blue and white seeds were retained in the remaining seed fractions, and only light blue seeds were blown out. The test was performed as follows: mixed seed batches were sorted into blue and white fractions, and a light blue blow-out procedure was performed on the (blown) blue fraction. To test the efficiency of this procedure, white seeds were manually removed from the retained blue fraction to obtain true blue video spectrometer values, and two complete culture dishes of that fraction were analyzed by video spectrometer to obtain the average value for each dish. The average of the two measurements was compared to the average video spectrometer values ​​of the two complete culture dishes of the blown-out fraction. The retained blue values ​​were significantly lower than those of the blown-out fraction, indicating that light blue (1n and 2n blue) seeds were preferentially blown out. Blowing was performed using different Bla strains and different backlight HSL settings in such a way that the target amount of light blue blown out (estimated at approximately 30% based on known breeding runs performed on these seed batches) was adjusted only by increasing the pixel size (starting from 600) in case of overblowing or decreasing the pixel size of the object setting in case of underblowing. This was tested on 6 seed batches (3 lighter and 3 darker blue aleurone layer seed batches). The HSL backlight and pixel settings used here to provide good light blue seed output are: H71 / S255 / L123, pixel size 350-450 (e.g., pixel size 400); H0 / S255 / L110, pixel size from 550 to 650 (e.g., pixel size 600); H20 / S255 / L150, pixel size from 550 to 650 (e.g., pixel size 600); and H47 / S255 / L123, pixel size from 650 to 750 (e.g., pixel size 700). White and cyan backlights (pixel size reduced to 300) do not provide any good light blue output—even at such low pixel sizes, the amount of blue seed output is very small, and those seeds contain white seeds, resulting in very bright video spectrometer values ​​(not shown below). See Table 9, which indicates the LB blow video spectrometer values ​​for three light blue (L) or dark blue (D) Bla seed batches (backlit H20 / S255 / L150), and Table 10, which shows the different backlit colors tested for the dark blue D3 strain.

[0371] In this LB blowing method, the L-type light blue seeds to be blown out can be used. min The object setting is: 97 ≤ L min ≤ 110, such as L min It is 105, and S minThe object can be set to: 65 ≤ S min ≤ 75, such as 70 for S min , and L max It could be: 138-145, such as L max It is 140, and S max It can be: 95-255, such as S max It is 120.

[0372] The color setting of the objects to be sorted can be 345 ≤ H ≤ 45 or 344 ≤ H ≤ 45, or it can be any H range, as long as it includes the range from H ≥ 354 and H ≤ 45 (i.e., including 354 ≤ H ≤ 45), such as the entire H range (e.g., from 345 to 344).

[0373] Table 9

[0374]

[0375] Table 10

[0376]

[0377] The brightness of the blue seeds blown out is due to the appearance of white seeds during the blowing process.

[0378] Example 7: Identifying diseased seeds with dark spots from uninfected wheat seeds.

[0379] In this example, 100 grams of wheat seed fraction infected with ergot fungus (containing ergot fruiting bodies and dark spots / dark seeds, interspersed with uninfected seeds) was mixed with a 300-gram batch of uninfected wheat seeds from a hybrid. Before mixing, the video spectrometer values ​​of two intact culture dishes for both fractions were measured (ergot-infected fraction: video spectrometer value 0.95, clean seed fraction: video spectrometer value 1.48). After mixing, the same procedure was performed on the resulting mixed fraction (mixed fraction video spectrometer value: 1.18). The ergot-infected seed fraction had the lowest average value, while the average value of the mixed fraction fell between the (dark) value of the ergot-infected fraction and the (bright) value of the clean seed fraction. To test the effect of different backlight colors on the effectiveness of removing ergot fruiting bodies and dark seeds from clean seeds using a dark-blowing procedure (low L / brightness method, HSL object settings: H ≥ 354 H ≤ 54, Lmin: 0, Lmax: 110, Smin: 50, Smax: 90, pixel size 200 (to cover small dark spots)), a sorting / blowing was performed on an ASM® EUREKA sorter, and the blown-out and retained fractions were examined using a video spectrometer. The values ​​obtained are the average of those examined from two complete culture dishes.

[0380] Effective blowout was characterized by the dark blown-out fraction (low video spectrometer value) and the light retained fraction. The value of the retained fraction was almost as high as the reference value for clean seeds (1.48), indicating efficient removal of ergot fruiting bodies and dark, colored seeds. Such results were achieved only with H71 (green) and H20 (orange) backlight colors, and almost achieved with H0 (red). For the latter, it should be noted that the Lmax value could have been slightly increased to blow out 77g (as in the case of H71 green, and for H20 orange, Lmax increased to 117), instead of just 61g, for better results.

[0381] However, despite higher yields, the brightness of the retained fractions produced by blowing with cyan or white backlighting (1.33 and 1.38) was significantly lower than the reference value, and these fractions were also less visually clean (still containing some fruiting body fragments and colored dark seeds). This result (see Table 11) clearly demonstrates the superior performance of red, green, or orange backlighting compared to standard cyan or white backlighting in removing ergot and / or dark-infected seeds.

[0382] Table 11

[0383]

[0384] Example 7: Sorting out darker red rice varieties from red rice samples

[0385] A preferred backlight HSL setting, featuring a low-L object setting such as that used when sorting deep blue aleurone layer seeds, was also tested to obtain a more uniform batch of rice seeds on 500g of commercially packaged red rice grains (with lighter and darker red seed colors). A certain percentage (approximately 12%) of the darkest seeds (containing more colorants / pigments, such as anthocyanins and / or proanthocyanidins) was blown out of the red rice batch using a brightness (low-L) method to achieve a more uniform appearance. The procedural settings were derived from images of red rice grains taken in an ASM® EUREKA sorter and adjusted so that all grains in the images could be identified by these settings. Then, by adjusting the L... min Change L to 0. max The procedure was modified to produce approximately 60 grams of the deepest red seeds. To compare the efficiency of various backlight color / HSL settings, video spectrometer values ​​for the blown fraction and the resulting / retained fraction were measured (given as an average of two culture dishes (approximately 620 seeds per dish)). At the same weight, lower blown video spectrometer values, or higher values ​​for the resulting / retained fractions respectively, indicate a better selection of backlight for this purpose.

[0386] The results are shown in Table 12 below.

[0387]

[0388] Compared to cyan or white backlighting, backlighting with H71, H111, H0, or H20 all resulted in improved sorting of deep red rice, producing a more uniform overall red rice sample with a slightly lighter appearance.

[0389] The backlight in the table above, indicated only by the H value, is... Figures 4 to 9 The ones described in the text are H71 / S255 / L123 or H20 / S255 / L150 (cyan backlight is H180 / S255 / L128 and white backlight is H0 / S255 / L255).

[0390] List of reference numerals

[0391] 110 sorting equipment

[0392] 112 Grain Seeds

[0393] 114 colorless grain seeds

[0394] 116 seeds containing a blue aleurone layer

[0395] 118 Seed Feeder

[0396] 120 seed stream

[0397] 122 sorting station

[0398] 124 feed hopper

[0399] 126 chute

[0400] 128 Vibrating Feeder

[0401] 130 backlight device

[0402] 131 Front Lighting Equipment

[0403] 132 camera

[0404] 134 First Backlight Device

[0405] 136 First Camera

[0406] 138 First Side

[0407] 140 Second Side

[0408] 142 Second backlight device

[0409] 144 Second Camera

[0410] 146 Injector

[0411] 148 pneumatic injector

[0412] 150 target chute

[0413] 152 sorting chute

[0414] 154 controller

[0415] 156 supply seed stream

[0416] 158 images

[0417] 160 Identify the seeds to be sorted

[0418] 162 sprays of the identified seeds to be sorted

[0419] 164 Angles around the center

[0420] 166 Distance from the center

[0421] 168 frames

[0422] 170x axis

[0423] 172y axis

[0424] Mixed fraction of 200 grain seeds

[0425] 202 First Backlight

[0426] 204 seeds were sprayed out of the grade.

[0427] The retained grades of 206 seeds

[0428] 208 Second Backlight

[0429] 210 First Round of Sorting

[0430] 212 Second Round of Sorting

[0431] 214 Waste Classification of Grain Seeds

[0432] 216 Third Round of Sorting

[0433] 218 Fourth Round of Sorting

[0434] 220 Fourth Backlight

[0435] 222 Third backlight.

Claims

1. A method for sorting cereal seeds (112), wherein the cereal seeds (112) comprise colorless cereal seeds (114) and seeds containing a blue aleurone layer (116), or comprise darker seeds containing a blue aleurone layer and lighter seeds containing a blue aleurone layer, the method comprising: i. Supplying a seed stream (120) to a sorting station (122) including at least one backlighting device (130) for backlighting the seeds (112) of the seed stream (120) and at least one camera (132) for taking at least one image of the backlit seeds. ii. Using the camera (132), take at least one image of the seeds of the seed stream (120) backlit; iii. Automatically identify the seeds to be sorted from the seed stream (120) from the images captured in step ii.; as well as iv. Automatically eject the identified seeds to be sorted from the seed stream (120), The at least one backlight has an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space.

2. The method according to the preceding claim, wherein in step iii., the seeds to be sorted from the seed stream (120) are identified by recognizing objects in the image that cumulatively satisfy the following conditions: These objects have predefined color coordinates, specifically color coordinates within a predefined range in the HSL color space, and These objects have one or more of the following: predefined area, predefined size, predefined diameter, predefined equivalent diameter, and predefined shape.

3. The method according to any one of the preceding claims, wherein the at least one backlight has an H coordinate in the HSL color space that is within at least one range selected from the group consisting of: -0 ≤ H < 25 or H ≥ 310, specifically excluding the range 332 ≤ H ≤ 338; -70 ≤ H ≤ 150； -25 ≤ H ≤ 70。 4. The method according to any one of the preceding claims, wherein the at least one backlight has an S-coordinate in the HSL color space of 0.25 ≤ S ≤ 1.0, specifically 0.5 ≤ S ≤ 1.0, and more specifically 0.75 ≤ S 1.

0.

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

6. The method according to any one of the preceding claims, wherein, The seeds to be separated from the seed stream (120) are those containing a blue aleurone layer (116).

7. The method according to any one 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 image stream of the seed stream (120) is captured, and wherein in step iii, the image stream is continuously evaluated to continuously identify the seeds to be sorted from the seed stream (120).

8. The method according to any one of the preceding claims, wherein a batch of seeds is provided, and wherein method steps i.-iv. are repeated on the batch of seeds, wherein, In each repetition, the batch is reduced due to the seeds sprayed in step iv. of the previous run.

9. 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 spectral piece of information, such as near-infrared spectral information, about the seeds (112) in the seed stream (120), wherein the at least one spectral piece of information is used to automatically identify the seeds to be sorted from the seed stream (120).

10. The method according to any one of the preceding claims, wherein the method comprises performing steps i. to iv. using at least one first backlight (202) having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the method further comprises repeating steps i. to iv. using at least one second backlight (208) for one of the ejected fraction (204) of the seed or the retained fraction (206) of the seed, the second backlight (208) having an H coordinate of H ≤ 150 or H ≥ 310 and an L coordinate of 0.10 ≤ L ≤ 0.80 in the HSL color space, wherein the second backlight (208) is different from the first backlight (202).

11. The method according to the preceding claim, wherein the method comprises repeating steps i. to iv. on the ejected fraction (204) and the retained fraction (206) of the seed, respectively.

12. The method according to any one of the preceding claims, for sorting blue aleurone seeds from a seed mixture comprising darker and lighter blue aleurone seeds (116) and colorless seeds (114), wherein the sorting procedure uses a combination of the following two: a) an HSL object setting for the dark blue aleurone seeds (116) provided by the sorting device (110), and b) an HSL object setting for the light blue aleurone seeds (116) provided by the sorting device (110), for example by sorting out: (1) objects with H coordinates of 5 ≤ H ≤ 65, S coordinates of 18 ≤ S ≤ 82, L coordinates of 88 ≤ L ≤ 145 and dimensions of 200-1500, or 400-1000, or at least 700 pixels, and (2) objects with H coordinates of 19 ≤ H ≤ 79, S coordinates of 19 ≤ S ≤ 79, and dimensions ... 83 and an object with an L coordinate of 62 ≤ L ≤ 130 and a size of 200-1500, or 400-1000, or at least 700 pixels, preferably using an HSL backlight setting with a hue value of 340, 71, or 0, the method may be performed once or may be repeated multiple times, such as 1-3 times, to maximize the purity of the colorless seed, or in the method, the object in the image has an H coordinate in (1) and (2) that includes the range of 354 ≤ H ≤ 45.

13. The method according to any one of claims 1 to 11, for separating dark blue 3n aleurone layer seeds from a seed mixture comprising dark blue 3n aleurone layer seeds and light blue 1n or 2n aleurone layer seeds, the mixture may contain colorless seeds, wherein L min The object setting is lowered to at least the value of the darkest seed to ensure that the darkest blue seed is blown out, thereby modifying the L object setting provided by the sorting device (110) for the dark blue 3n aleurone layer seed, and wherein L max The value is set to the darkest seed that causes the desired weight percentage to be blown out, and the H object setting is any H range including 354 ≤ H ≤ 45, and in the method, the backlight is as described in any of the preceding claims, or a white LED backlight or a backlight with an L coordinate of L = 255 (white).

14. The method according to the preceding claim, wherein, The amount of dark blue 3n aleurone layer seeds in a batch of blue aleurone layer seeds (116) was reduced by decreasing the L of the scheme. min Object settings to reduce, such as L to 0 min or 0 ≤ L min ≤ 60 or 0 ≤ L min ≤ 30, such as 78 ≤ L max L ≤ 127 max L max The value depends on the percentage or amount of dark blue 3n aleurone layer seeds to be removed and on the color depth of the seed batch.

15. A sorting device (110) for sorting grain seeds (112), comprising: I. At least one seed feeder (118) for supplying a seed stream (120) to at least one sorting station (122). as well as II. The at least one sorting station (122) includes at least one backlighting device (130) for backlighting the seeds of the seed stream (120), the sorting station (122) further includes at least one camera (132) for capturing at least one image of the backlit seeds, and the sorting station (122) further includes at least one ejector (146) for ejecting seeds from the seed stream (120). The sorting device (110) is configured to perform the method according to any one of the preceding claims.

16. A computer program comprising instructions that, when executed by a sorting device (110) according to any one of the preceding claims relating to a sorting device (110), cause the sorting device (110) to perform the method according to any one of the preceding claims relating to a method.

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

18. The sorting device (110) according to any one of the preceding claims relating to the sorting device (110) is used for the purpose selected from the group consisting of: sorting colored seeds from a seed mixture containing colorless seeds and colored seeds; sorting colored grain seeds from a grain seed mixture (112) containing colorless grain seeds (114) and colored grain seeds; sorting colored grain seeds (116) containing a blue aleurone layer from a grain seed mixture (112) containing colorless grain seeds (114) and grain seeds (116) containing a blue aleurone layer; sorting colored grain seeds (116) containing 1n, 2n and 3n blue aleurone layers. Colored grain seeds (116) containing a deep blue 3n aleurone layer are sorted from a mixture of grain seeds (116) containing a white layer; this mixture may contain colorless grain seeds (114); colored grain seeds (116) containing a blue aleurone layer are sorted from a mixture of grain seeds (116) containing a blue aleurone layer and colorless grain seeds (114); lighter-colored grain seeds (116) containing a 1n or 2n blue aleurone layer are sorted from a mixture of grain seeds (116) containing 1n, 2n and 3n blue aleurone layers; this mixture may contain colorless grain seeds (114); using a color space with 70 Using at least one backlight with an H coordinate of ≤ H ≤ 150, colored grain seeds (116) containing a blue aleurone layer are sorted from a mixture of grain seeds (112) containing colorless grain seeds (114) and grain seeds (116) containing a blue aleurone layer; using at least one backlight with an H coordinate of 25 ≤ H ≤ 70 in the HSL color space, colored grain seeds (116) containing a blue aleurone layer are sorted from a mixture of grain seeds (112) containing colorless grain seeds (114) and ...

Images