Evaluation of germinating seedlings

Abstract

There is disclosed an image processing method for assessing a germinating seedling. The method comprises: identifying at least a part of a germinating seedling within an image; and determining one or more geometric characteristics of the identified at least part of the germinating seedling. The method may further comprise determining one or more parameters for assessing the germinating seedling based on the one or more geometric characteristics. The one or more geometric characteristics may comprise one or more of: a length; and a surface area of the at least part of the germinating seedling. At least part of the germinating seedling may comprise one or more of: a root; a shoot; a seed; and a leaf of the at least part of the germinating seedling.

Description

[0001] EVALUATION OF GERMINATING SEEDLINGS

[0002] BACKGROUND

[0003] Field

[0004] Certain examples and embodiments of the present invention provide one or more techniques for evaluating germinating seedlings.

[0005] Description of the Related Art

[0006] It is often important to measure the physiological quality of seedlings, for example, vigour (i.e. the ability of the seed to withstand stress such as low or high temperatures) and / or the rate of emergence during germination. Such an assessment can be used, for example, to measure the quality of a certain strain, to compare different strains and / or to determine or compare the effects (positive or negative) of one or more seed treatments on one or more strains.

[0007] Typically, the quality of a seedling is judged by a person through visual inspection. The seedling is then typically categorised as ‘normal seedling’ or ‘abnormal seedling’ or ‘dead seed’. One problem with this approach is that the person judging the seedling quality needs to be sufficiently well trained. Another problem is that the assessment can be reatively time consuming, particularly when assessing a relatively large number of seedlings. Furthermore, the categorisation of ‘normal’ and ‘abnormal’ only provides a relatively coarse and / or imprecise measure of quality.

[0008] The above information is presented as background information only to assist with an understanding of the present invention. No determination has been made, and no assertion is made, as to whether any of the above might be applicable as prior art with regard to the present invention.

[0009] SUMMARY

[0010] It is an aim of certain examples and embodiments of the present invention to address, solve, mitigate or obviate, at least partly, at least one of the problems and / or disadvantages associated with the related art, for example at least one of the problems and / or disadvantages mentioned herein. Certain examples and embodiments of the present invention aim to provide at least one advantage over the related art, for example at least one of the advantages mentioned herein.

[0011] The present invention is defined in the independent claims. Advantageous features are defined in the dependent claims.

[0012] Certain examples of the present invention provide an image processing method for assessing a germinating seedling, the method comprising: identifying at least a part of one or more germinating seedling, and preferably at least a part of two or more germinating seedlings, within an image; and determining one or more geometric characteristics of the identified at least part of the one or more germinating seedlings, and preferably of the identified at least part of the two or more germinating seedlings.

[0013] In certain examples, the method may further comprise determining one or more parameters for assessing the one or more germinating seedlings, and preferably for assessing the two or more germinating seedlings, based on the one or more geometric characteristics.

[0014] In certain examples, the one or more parameters may comprise one or more of: a value corresponding to one geometric characteristic; and a value calculated based on two or more geometric characteristics.

[0015] In certain examples, the one or more geometric characteristics may comprise one or more of: a length; and a surface area of the at least part of the one or more germinating seedlings, and preferably the at least part of the two or more germinating seedlings.

[0016] In certain examples, the at least part of the one or more germinating seedlings, and preferably the at least part of the two or more germinating seedlings, may comprise one or more of: a root; a shoot; a seed; and a leaf.

[0017] In certain examples, identifying the at least part of the one or more germinating seedlings, and preferably the at least part of the two or more germinating seedlings, may be performed based on one or more of: segmentation; filtering; edge detection; and a machine learning algorithm.

[0018] In certain examples, the filtering may be performed based on one or more of: hue; brightness; saturation; shade; and colour of the at least part of the one or more germinating seedlings, and preferably of the at least part of the two or more germinating seedlings.

[0019] In certain examples, the at least part of the one or more germinating seedlings, and preferably the at least part of the two or more germinating seedlings, may be identified using deep learning semantic segmentation model and combining it with an image processing technique to classify each pixel within an image of a seedling, wherein classifying may comprise determining if each pixel relates to at least one of seed, root, shoot or background.

[0020] In certain examples, the method may further comprise performing a calibration procedure for calibrating the one or more geometric characteristics to one or more predetermined units of measurement.

[0021] In certain examples, the calibration procedure may comprise: identifying an object of known size in an image; determining at least one geometric characteristic of the object; and determining a conversion factor based on the known size of the object and the determined geometric characteristic of the object.

[0022] In certain examples, the method may further comprise obtaining the image based on one or more of: capturing the image using a camera; receiving the image from an external electronic device; and retrieving a stored image.

[0023] In certain examples, the one or more geometric characteristics may comprise one or more of: one or more geometric characteristics based on a single germinating seedling; one or more cumulative geometric characteristics based on two or more germinating seedlings; and one or more average geometric characteristics based on two or more germinating seedlings.

[0024] In certain examples, the method may further comprise determining the number of germinating seedlings in the image.

[0025] Certain examples of the present invention provide an apparatus for assessing a germinating seedling, the apparatus comprising a processor configured to: identify at least a part of one or more germinating seedling, and preferably at least a part of two or more germinating seedlings, within an image; and determine one or more geometric characteristics of the identified at least part of the one or more germinating seedlings, and preferably of the identified at least part of the two or more germinating seedlings.

[0026] Certain examples of the present invention provide a non-transitory computer-readable medium comprising instructions that, when executed, cause a processor of a computing apparatus to identify at least a part of one or more germinating seedling, and preferbaly at least a part of two or more germinating seedlings, within an image; and determine one or more geometric characteristics of the identified at least part of the one or more germinating seedling, and preferbaly the identified at least part of the two or more germinating seedlings, wherein the one or more geometric characteristics comprise one or more of: one or more cumulative geometric characteristics based on the two or more germinating seedlings; and one or more average geometric characteristics based on the two or more germinating seedlings.

[0027] The term “seedling” herein relates to a seedling of a crop or plant species including but not limited to cereals, such as wheat, barley, rye, oats, rice, maize (e.g. field corn, popcorn, corn), millet or sorghum; beet, such as sugar or fodder beet; fruit, for example pomaceous fruit, stone fruit or soft fruit, such as apples, pears, plums, peaches, almonds, cherries or berries, for example strawberries, raspberries or blackberries; leguminous crops, such as beans, lentils, peas or soya (soya beans); oil crops, such as oilseed rape, mustard, poppies, olives, sunflowers, coconut, castor, cocoa or ground nuts; cucurbits, such as pumpkins, cucumbers, melons, watermelons or squashes; fibre plants, such as cotton, flax, hemp or jute; citrus fruit, such as oranges, lemons, grapefruit or tangerines; vegetables, such as spinach, lettuce, asparagus, cabbages, broccolis, cauliflowers, carrots, onions, tomatoes, potatoes, eggplants, peppers or bell peppers; Lauraceae, such as avocado, Cinnamonium or camphor; and also tobacco, nuts, coffee, sugarcane, tea, grapevines, hops, the plantain family and latex plant. In a preferred embodiment, the plant can be selected from:

[0028] - cereals, such as wheat, barley, maize (e.g. field corn, popcorn, corn);

[0029] - leguminous crops, such as peas or soya (soya beans); and

[0030] - oil crops, such as sunflower.

[0031] In certain examples, the apparatus may further comprise an imaging device configured to capture the image.

[0032] Certain examples of the present invention provide a computer program comprising instructions which, when the program is executed by a computer or processor, cause the computer or processor to carry out a method according to any example, embodiment, aspect and / or claim disclosed herein.

[0033] Certain examples of the present invention provide a computer or processor-readable data carrier having stored thereon a computer program according to the preceding example.

[0034] Embodiments, aspects or examples disclosed in the description and / or figures falling outside the scope of the claims are to be understood as examples useful for understanding the present invention.

[0035] Other aspects, advantages, and salient features of the present invention will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the accompanying drawings, disclose examples and embodiments of the present invention.

[0036] BRIEF DESCRIPTION OF THE FIGURES

[0037] Figure 1 illustrates an exemplary method for evaluating germinating seedlings using image processing;

[0038] Figure 2 illustrates an exemplary system for evaluating germinating seedlings using image processing;

[0039] Figure 3a is a first example image of a number of germinating seedlings (corn), Figure 3b is a processed image showing the result of image processing to identify the seedlings in the image of Figure 3a, and Figure 3c is a processed image showing the result of image processing to identify different parts of the seedlings in the image of Figure 4a;

[0040] Figure 4a is a second example image of a number of germinating seedlings (sunflower), Figure 4b is a processed image showing the result of image processing to identify the seedlings in the image of Figure 4a, and Figure 4c is a processed image showing the result of image processing to identify different parts of the seedlings in the image of Figure 4a;

[0041] Figures 5a and 5b are tables showing exemplary results of evaluating germinating seedlings, including those of Figures 3a-3c and 4a-4c, using image processing with comparison to the conventional technique;

[0042] Figures 6a and 6b are example processed images of a number of germinating corn seedlings subject to treatment CO (Figure 6a) and treatment C5 (Figure 6b); and

[0043] Figures 7a and 7b are example processed images of a number of germinating sunflower seedlings subject to treatment S1 (Figure 7a) and treatment S3 (Figure 7b).

[0044] DETAILED DESCRIPTION

[0045] The following description of examples and embodiments of the present invention, with reference to the accompanying drawings, is provided to assist in a comprehensive understanding of the present invention, as defined by the claims. The description includes various specific details to assist in that understanding but these are to be regarded as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the examples described herein can be made.

[0046] According to various examples and embodiments of the present invention, germinating seedlings may be evaluated using image processing. According to various examples and embodiments of the present invention, the seedlings may be evaluated based on one or more geometric characteristics of the seedlings determined based on the image processing. Various non-limiting examples are described in more detail below.

[0047] Figure 1 illustrates an exemplary method for evaluating germinating seedlings using image processing.

[0048] Referring to Figure 1 , in a first operation 101 , an image of one or more germinating seedlings is obtained.

[0049] For example, the image may be obtained by capturing a image of one or more seedlings using an imaging device such as a camera. Alternatively, the image may be obtained by retrieving an existing image stored in a database or other internal or external source.

[0050] In a second operation 102, the one or more seedlings are identified in the image using image processing.

[0051] Any suitable image processing techniques may be used to identify the seedlings, for example including segmentation, filtering, edge detection and / or processing using Artificial Intelligence (Al). For example, filtering may be performed based on hue, brightness, saturation and / or any other suitable visual charcateristic. In one example, if the seedlings are displayed against a relatively light background then parts of the image that are relatively dark (e.g. pixels having a brighteness less than a certain threshold) may be identified as the seedlings. In another example, if the background is a known colour, being a colour that differs from colours typically appearing in seedlings, then the background may be identified and filtered out based on the known colour, so that the remaining parts of the image may be identified as the seedlings. In a further example, parts of the image having colours typically associated with seedlings may be identified as the seedlings. In a yet further example, an edge detection filter may be applied to the image to identify the edges of the seedlings and parts of the image falling inside closed regions may be identified as the seedlings. In a still further example, the image may be segmented using any suitable algorithm such as the watershed algorithm and the seedlings may be identified as one or more resulting segments. In another example, a trained Al Machine Learning (ML) engine may be used to identify the seedlings.

[0052] In some examples, there may be only one seedling in the image. In other examples, there may be more that one seedling in the image. In the latter case, in certain examples the image processing may identify parts (i.e. pixels) of the image corresponding to seedlings without distinguishing between different seedlings. However, in other examples the image processing may be capable of distinguishing between individual seedlings and may identify different parts of the image corresponding to respective different seedlings.

[0053] Figure 3a is a first example image of a number of germinating seedlings (corn) and Figure 3b is a processed image showing the result of image processing to identify the seedlings in the image of Figure 3a. Figure 4a is a second example image of a number of germinating seedlings (sunflower) and Figure 4b is a processed image showing the result of image processing to identify the seedlings in the image of Figure 4a. The images of Figures 3a and 4a include multiple seedlings. The areas of the images of Figures 3a and 4a corresponding to the seedlings are shaded gray against a light background in Figures 3b and 4b.

[0054] An individual seedling comprises different parts, for example including roots, shoots, seed and leaves. In certain examples, the image processing may identify parts (i.e. pixels) of the image corresponding to a seedling without distinguishing between different parts of the seedling. However, in other examples the image processing may be capable of distinguishing between different parts of the seedlings and may identify different parts of the image corresponding to respective different parts of the seedlings. For example, different parts of a seedling may be characterised by different shades and / or colours, in which case the image processing may be able to identify different parts of the seedling based on shade and / or colour. In certain examples, a trained AI / ML engine may be used to identify different parts of the seedlings.

[0055] Figure 3c is a processed image showing the result of image processing to identify different parts of the seedlings in the first example image of Figure 3a, and Figure 4c is a processed image showing the result of image processing to identify different parts of the seedlings in the second example image of Figure 4a. The areas of the images of Figures 3a and 4a corresponding to different parts of the seedlings are shaded with different colours / grayscale levels in Figures 3c and 4c. In Figures 3c and 4c, the upper part of the seedlings are the shoots and the lower part of the seedlings are the roots.

[0056] Referring back to Figure 1 , in a third operation 103, one or more geometric characteristics of the identified seedling(s) and / or one or more parts thereof are determined. For example, the geometric characterics may include a length, width and / or a cross sectional area of the identified seedling(s) and / or one or more parts thereof. The cross sectional area may refer, for example, to the apparent cross sectional area as observed in the view plane, i.e. the area in the image occupied by the seedling(s) and / or part(s) thereof.

[0057] The area may be determined, for example, by counting the number of pixels belonging to one or more areas corresponding to the identified seedling(s) and / or one or more parts thereof. The length may be determined, for example, by identifying a coordinate value (e.g. y-coordinate) of an upper point of the seedling or part thereof, identifying a coordinate value of a lower point of the seedling or part thereof, and determining the difference, in pixels, between the coordinates of the upper and lower points. Typically, a seedlings does not lie in an exact straight line. In this case, the length may be determined as the path length along a seedling. However, this approach may be more computationally expensive.

[0058] The geometric characteristic (e.g. length or area) may be determined with respect to all seedlings in the image as a group. For example, the total area of all seedlings may be determined. Similarly, the overall length of a group of seedlings may be determined (e.g. based on identifying coordinates of upper and lower points of the overall group). Alternatively, if individual seedlings can be distinguished in the image then the geometric characteristic may be determined with respect to an individual seedling. For example, the area and / or length of an individual seedling may be determined.

[0059] The geometric characteristic (e.g. length or area) may be determined with respect to one or more entire seedlings and / or one or more particular parts (e.g. root or shoot) of one or more seedlings. For example, the total area of a whole seedling (or group of seedlings) may be determined. Alternatively, the total area of only the roots, or only the shoots, of a seedling (or group of seedlings) may be determined.

[0060] In the case where there are multiple seedlings in the image, the geometric characteristic may be determined as a cumulative amount and / or an average amount. For example the total area of all seedlings in the image may be determined. Alternatively, the average area (e.g. mean, median or any other suitable statistical measure) per seedling may be determined. Similarly, an average length per seedling may be determined.

[0061] To determine an average value per seedling it may be necessary to know the number of seedlings in the image. In some examples, the number of seedlings in the image may be known in advance. For example a set number of seedlings may be used when capturing the image. In other examples, the number of seedlings may be determined based on the image processing. In this case, the number of seedlings may not need to be known in advance and a variable number of seedlings may appear in different images.

[0062] In the examples described above, the geometric characteristics (e.g. length or area) are determined in units of pixels. The absolute size of a pixel will depend on various factors, such as the pixel resolution of the image capturing device and the distance between the image capturing device and the seedings. In certain examples, a calibration procedure may be performed to allow the geometric characteristics to be converted to any suitable unit of measurement, for example centimeters (cm) and square centimeters (cm2). Examples of calibration procedures will now be described.

[0063] In certain examples, the image including the seedlings may also include an object of known size in a certain unit of measurement. For example, the object may comprise a predetermined shape, or a scale or ruler displaying markers having a known separation (e.g. millimeter markers). The object is identified in the image using image processing, and the size of the object in units of pixels is determined. Then, a conversion factor between pixels and the unit of measurement may be calculated based on a ratio of the determined number of pixels and the known size of the object.

[0064] In some examples, a calibration procedure may be performed once in advance. For example an image of an object of known size may be captured and a conversion factor may be determined as described above. After determining the conversion factor, images of the seedlings may be captured without the object of known size. Assuming the factors affecting the conversion factor (e.g. the distance between the imaging device and the captred subject) remain the same, then the same conversion factor may be applied to the images of the seedlings.

[0065] In certain example a calibration procedure may not be required. For example, determining the geometric characteristic(s) in units of pixels may be sufficient in some implementations, for example when performing relative comparison between different results.

[0066] Referring back to Figure 1 , in a fourth operation 104, one or more parameters are determined based on the one or more geometric characteristics. The parameter(s) may be used for example to assess the quality of the seedling(s). In some examples, a parameter may simply be a geometric characteristic, for example area or length. In other examples, a parameter may be based on two or more geometric characteristics, for example based on a weighted sum or any other suitable function. For example, a parameter may be calculated based on a predetermined function of both area and length and / or a predetermined function of the lengths of two different parts of a seedling.

[0067] Figure 2 illustrates an exemplary system 201 for evaluating germinating seedlings using image processing. The system 201 comprises an imaging device 203 such as a camera, a processor 205, a transceiver 207, and a display 209. The system of Figure 2 may be implemented in any suitable form. For example, the system may be provided in the form of a general purpose computer system executing dedicated software, as dedicated hardware, or as an application executing on a mobile device such as a mobile telephone or tablet.

[0068] The system 201 is configured to obtain an image, as described above in relation to operation 101 of Figure 1. For example, the camera 203 is configured to capture an image of one or more seedlings. The captured image may be transmitted to an external data source via the transceiver 207 for later use and / or for record keeping. Alternatively, the image may be obtained by retrieving a previously captured image from an internal memory, or from the external data source via the transceiver 207.

[0069] The processor 205 is configured to identify one or more seedlings in the obtained image using image processing, as described above in relation to operation 102 of Figure 1. The processor 205 is further configured to determine one or more geometric characteristics of the identified seedling(s) and / or one or more parts thereof, as described above in relation to operation 103 of Figure 1. The processor 205 is further configured to determine one or more parameters based on the one or more geometric characteristics, as described above in relation to operation 104 of Figure 1.

[0070] The system 201 maybe configured to output the determined parameter(s), for example on the display 209. The system 201 may be configured to transmit the determined parameter(s) via the transceiver 207 for storage in an external data source, and / or to store the parameter(s) in internal memory. In either case, the parameter(s) may be stored together or in association with the corresponding image.

[0071] The skilled person will appreciate that one or more of the components of the system illustrated in Figure 2 may be omitted in certain examples. For example, if the system obtains an image from an external data source or internal memory then the imaging device 203 may not be required. If the system obtains an image using the imaging device 203 and does not require communication with an external data source then the transceiver 207 may not be required. If the system is not required to display the determined parameter(s) to a user then the display 209 may not be required.

[0072] Figures 5a and 5b are tables showing exemplary results of evaluating germinating seedlings, including those of Figures 3a-3c and 4a-4c, using image processing with comparison to the conventional technique.

[0073] The table of Figure 5a shows the results of assessing the quality of corn seedlings (first column) of a variety “Glorius” (second column) when applying two different types of seed treatment “CO” and “C5” (third column). Each seed sample, CO and C5, was separated into eight replicas (A-H) of seedlings (fourth column) with 50 seedlings in each replica (eighth column). Figures 3a-3c show the results of replica C of treatment C5. In the table of Figure 5a, all of the replicas A-H have been evaluated using the conventional technique, while only half the replicas (specifically replicas A, C, E and G) have been evaluated using the techniques disclosed herein.

[0074] The fifth, sixth and seventh columns indicate the numbers of seedlings in each replica categorised as “normal seedling” or “abnormal seedling” or “dead seed” using the conventional manual approach. The resulting percentage of seedlings that successfully germinated (i.e. the percentage of the total number of seedlings categorised as “normal”) is indicated in the ninth column. Based on these results, the average percent of successful germination for treatment CO is 92% and the average percent of successful germination for treatment C5 is 94.25%. Therefore, using the conventional approach, it appears that there is no significant difference (a different of only 2.25 percentage points) in successful germination between the use of treatments CO and C5.

[0075] On the other hand, the tenth column indicates the total surface area (in cm2) of each replica of seedlings as determined using the techniques disclosed herein. Based on these results, the average (mean) surface area for treatment CO is 128.7 cm2and the average (mean) surface area for treatment C5 is 221.7 cm2. Accordingly, there is a significant difference in surface area between treatments CO and C5, indicating a significant difference between the use of treatments CO and C5.

[0076] The table of Figure 5b shows corresponding results of assessing the quality of sunflower seedlings of a variety “Onestar” when applying two different types of seed treatment “S1” and “S3”. Each seed sample, S1 and S3, was separated to to eight replicas (A-H) of seedlings with 50 seedlings in each replica. Figures 4a-4c show the results of replica A of treatment S1 . In the table of Figure 5b, all of the replicas A-H have been evaluated using the conventional technique, while only half the replicas (specifically replicas A, C, E and G) have been evaluated using the techniques disclosed herein.

[0077] Based on these results, the average percent of successful germination for treatment S1 is 93% and the average percent of successful germination for treatment S3 is 94.75%. Therefore, using the conventional approach, it appears that there is no significant difference (a different of only 1.75 percentage points) in successful germination between the use of treatments S1 and S3.

[0078] On the other hand, the average surface area for treatment S1 is 156.4 cm2and the average surface area for treatment S3 is 117.9 cm2. Accordingly, there is a significant difference in surface area between treatments S1 and S3, indicating a significant difference between the use of treatments S1 and S3.

[0079] The results in the tables of Figure 5a and Figure 5b thus demonstrate that the techniques disclosed herein are able to discriminate between the effectiveness of different treatments much more clearly and with a greater level of precision as compared to the conventional technique.

[0080] The results in the tables of Figures 5a and 5b may also be used to compare the effects of the different treatments on the roots and shoots of the seedlings. In relation to the table of Figure 5a, Figures 6a and 6b show the results of replica A for each of treatments CO (Figure 6a) and C5 (Figure 6b). In relation to the table of Figure 5b, Figures 7a and 7b show the results of replica A for each of treatments S1 (Figure 7a) and S3 (Figure 7b).

[0081] The following is based on the table of Figure 5a relating to corn seedlings. As noted above, using the conventional manual approach, the average percent of successful germination for treatment CO is 92% and the average percent of successful germination for treatment C5 is 94.25%. On the other hand, using the techniques disclosed herein, the following table gives the average (mean) root area (in pixels / R) and average (mean) shoot area (in pixels / R) for treatments CO and C5. Here, R is the image resolution equal to the image width in pixels multiplied by the image height in pixels. Accordingly, the root area and shoot area are each given in units of pixels divided by the image resolution, R.

[0082] It can be seen from the above table that the average areas of both roots and shoots is significantly lower using treatment CO compared to using treatment C5. Thus, one can conclude that treatment CO results in a lower germination quality compared to treatment C5, and that the lower germination quality affects both roots and shoots.

[0083] The following is based on the table of Figure 5b relating to sunflower seedlings. As noted above, using the conventional manual approach, the average percent of successful germination for treatment S1 is 93% and the average percent of successful germination for treatment S3 is 94.75%. On the other hand, using the techniques disclosed herein, the following table gives the average (mean) root area (in pixels / R) and average (mean) shoot area (in pixels / R) for treatments S1 and S3.

[0084] It can be seen from the above table that the average roots area is significantly lower using treatment S3 compared to using treatment S1. On the other hand, there is not a significant difference in the average shoots area between treatments S3 and S1. Thus, one can conclude that treatment S3 results in a lower germination quality compared to treatment S1 , but only for roots. The results in the tables of of Figures 5a and 5b thus demonstrate that the techniques disclosed herein are able to discriminate between the effectiveness of different treatments to different parts of seedlings, in contrast to the conventional technique.

[0085] The results in the tables of Figures 5a and 5b include results based on area (total area, shoots area, roots area). However, a similar evaluation may be performed based on length (total length, shoots length, roots area) instead of, or in addition to, area. The skilled person will appreciate that the techniques described herein may be based on any suitable geometric characteristic(s), for example area and / or length. The skilled person will also appreciate that the techniques described herein may be based on whole seedlings and / or different part(s) of seedlings, for example roots and / or shoots.

[0086] According to the techniques disclosed herein, the process of judging the quality of seedlings may be automated, thereby reducing the time required for analysis compared to the conventional manual process, and avoiding the need to train a human operator. Furthermore, the techniques disclosed herein may be used to determine one or more parameters providing a more precice and discriminating means for assessing seedling quality compared to the categorisation of ‘normal’ and ‘abnormal’ in the conventional process.

[0087] The terms and words used in this specification are not limited to the bibliographical meanings, but are merely used to enable a clear and consistent understanding of the present invention.

[0088] The same or similar components may be designated by the same or similar reference numerals, although they may be illustrated in different drawings.

[0089] Detailed descriptions of elements, features, components, structures, constructions, functions, operations, processes, characteristics, properties, integers and steps known in the art may be omitted for clarity and conciseness, and to avoid obscuring the subject matter of the present invention.

[0090] Throughout this specification, the words “comprises”, “includes”, “contains” and “has”, and variations of these words, for example “comprise” and “comprising”, means “including but not limited to”, and is not intended to (and does not) exclude other elements, features, components, structures, constructions, functions, operations, processes, characteristics, properties, integers, steps and / or groups thereof.

[0091] Throughout this specification, the singular forms “a”, “an” and “the” include plural referents unless the context dictates otherwise. For example, reference to “an object” includes reference to one or more of such objects.

[0092] Throughout this specification, language in the general form of “X for Y” (where Y is some action, process, function, activity, operation or step and X is some means for carrying out that action, process, function, activity, operation or step) encompasses means X adapted, configured or arranged specifically, but not exclusively, to do Y.

[0093] Elements, features, components, structures, constructions, functions, operations, processes, characteristics, properties, integers, steps and / or groups thereof described herein in conjunction with a particular aspect, embodiment, example or claim are to be understood to be applicable to any other aspect, embodiment, example or claim disclosed herein unless incompatible therewith.

[0094] It will be appreciated that examples and embodiments of the present invention can be realized in the form of hardware, software or any combination of hardware and software. Any such software may be stored in any suitable form of volatile or non-volatile storage device or medium, for example a ROM, RAM, memory chip, integrated circuit, or an optically or magnetically readable medium (e.g. CD, DVD, magnetic disk or magnetic tape).

[0095] Certain examples and embodiments of the present invention provide a computer program comprising instructions which, when the program is executed by a computer or processor, cause the computer or processor to carry out a method according to any example, embodiment, aspect and / or claim disclosed herein. Certain examples and embodiments of the present invention provide a computer or processor-readable data carrier having stored thereon such a computer program.

[0096] The techniques described herein may be implemented using any suitably configured apparatus and / or system. Such an apparatus and / or system may be configured to perform a method according to any aspect, embodiment, example or claim disclosed herein. Such an apparatus may comprise one or more elements, for example one or more of receivers, transmitters, transceivers, processors, controllers, modules, units, and the like, each element configured to perform one or more corresponding processes, operations and / or method steps for implementing the techniques described herein. For example, an operation / function of X may be performed by a module configured to perform X (or an X-module). An apparatus and / or one or more elements thereof may be implemented in the form of hardware, software, a virtualised function instantiated on an appropriate platform (e.g. on a cloud infrastructure), or any combination of these.

[0097] While the invention has been shown and described with reference to certain examples, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention, as defined by the appended claims.

Claims

Claims1. An image processing method for assessing germinating seedlings, the method comprising: identifying at least a part of two or more germinating seedlings within an image; and determining one or more geometric characteristics of the identified at least part of the two or more germinating seedlings, wherein the one or more geometric characteristics comprise one or more of: one or more cumulative geometric characteristics based on the two or more germinating seedlings; and one or more average geometric characteristics based on the two or more germinating seedlings.

2. A method according to claim 1 , further comprising determining one or more parameters for assessing the two or more germinating seedlings based on the one or more geometric characteristics.

3. A method according to claim 2, wherein the one or more parameters comprise one or more of: a value corresponding to one geometric characteristic; and a value calculated based on two or more geometric characteristics.

4. A method according to claim 1, 2 or 3, wherein the one or more geometric characteristics comprise one or more of: a length; and a surface area of the at least part of the two or more germinating seedlings.

5. A method according to any preceding claim, wherein the at least part of the two or more germinating seedlings comprises one or more of: a root; a shoot; a seed; and a leaf.

6. A method according to any preceding claim, wherein identifying the at least part of the two or more germinating seedlings is performed based on one or more of: segmentation; filtering;edge detection; and a machine learning algorithm.

7. A method according to claim 6, wherein the filtering is performed based on one or more of: hue; brightness; saturation; shade; and colour of the at least part of the two or more germinating seedlings.

8. A method according to any preceding claim, further comprising performing a calibration procedure for calibrating the one or more geometric characteristics to one or more predetermined units of measurement.

9. A method according to claim 8, wherein the calibration procedure comprises: identifying an object of known size in an image; determining at least one geometric characteristic of the object; and determining a conversion factor based on the known size of the object and the determined geometric characteristic of the object.

10. A method according to any preceding claim, further comprising obtaining the image based on one or more of: capturing the image using a camera; receiving the image from an external electronic device; and retrieving a stored image.

11. A method according to any preceding claim, further comprising determining the number of germinating seedlings in the image.

12. An apparatus for assessing germinating seedlings, the apparatus comprising a processor configured to: identify at least a part of two or more germinating seedlings within an image; and determine one or more geometric characteristics of the identified at least part of the two or more germinating seedlings, wherein the one or more geometric characteristics comprise one or more of: one or more cumulative geometric characteristics based on the two or more germinating seedlings; andone or more average geometric characteristics based on the two or more germinating seedlings.

13. An apparatus according to claim 12, further comprising an imaging device configured to capture the image.

14. A computer program comprising instructions which, when the program is executed by a computer or processor, cause the computer or processor to carry out a method according to any of claims 1 to 11.

15. A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor of a computing apparatus to: identify at least a part of two or more germinating seedlings within an image; and determine one or more geometric characteristics of the identified at least part of the two or more germinating seedlings, wherein the one or more geometric characteristics comprise one or more of: one or more cumulative geometric characteristics based on the two or more germinating seedlings; and one or more average geometric characteristics based on the two or more germinating seedlings.

Images