Method for determining sensitivity of a crop to environmental stresses

Abstract

Disclosed is a computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising providing aerial image data based on one or more aerial images of a crop plot; calculating, based on aerial image data, a stunting component, a foliar damage component, and crop injury; calculating the crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury.

Description

[0001] Method for determining sensitivity of a crop to environmental stresses

[0002] TECHNICAL FIELD

[0003] Disclosed are a method for determining sensitivity of a crop to environmental stresses, a use, a system, a computer program product, and a computer-readable medium.

[0004] TECHNICAL BACKGROUND

[0005] Crop sensitivity, in general, can be estimated by looking into the damage to a crop that will be caused by certain stresses. Studying the sensitivity allows for selecting crops for planting, for breeding, e.g. to select particular traits, and / or selecting and / or developing treatments.

[0006] However, currently known methods are rather subjective. This can make them error-prone, difficult to reproduce, and potentially imprecise. Accordingly, there is need for improvement of known methods.

[0007] It is therefore an object of the present disclosure to provide an improved method for determining sensitivity of a crop to environmental stresses.

[0008] SUMMARY OF THE INVENTION

[0009] The present disclosure provides a computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising providing aerial image data based on one or more aerial images of a crop plot; calculating, based on the aerial image data, a stunting component and a foliar damage component (also referred to as a discoloration component in the present disclosure); calculating a crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury.

[0010] In other words, the present disclosure provides a computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising providing aerial image data of a crop plot; calculating, based on the aerial image data, a stunting component and a foliar damage component; calculating the crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury. Aerial image data of a crop plot may be image data that are based on one or more aerial images of the crop plot.

[0011] In yet other words, the present disclosure provides a computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising calculating, based on aerial image data of a crop plot, a stunting component and a foliar damage component; calculating a crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury. Aerial image data of a crop plot may be image data that are based on one or more aerial images of the crop plot.

[0012] The present disclosure provides a computer-implemented method for calculating a crop injury, the method comprising calculating, based on aerial image data of a crop plot, a stunting component and a foliar damage component; and calculating a crop injury through a weighted summation of the stunting component and the foliar damage component. Aerial image data of a crop plot may be image data that are based on one or more aerial images of the crop plot.

[0013] The present disclosure further provides a system comprising a computing system configured to carry out the method of the present disclosure.

[0014] In other words, the present disclosure provides a system comprising a computer system, the computer system configured to determine sensitivity of a crop to environmental stresses, wherein the determining comprises providing aerial image data based on one or more aerial images of a crop plot; calculating, based on the aerial image data, a stunting component and a foliar damage component; calculating a crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury.

[0015] In particular, the computing system may determine sensitivity of a crop to environmental stresses by executing a first module, a second module, a third module, and a fourth module, wherein the first module is configured to provide aerial image data based on one or more aerial images of a crop plot, wherein the second module is configured to calculate, based on the aerial image data, a stunting component, and a foliar damage component, wherein the third module is configured to calculate crop injury through a weighted summation of the stunting component and the foliar damage component, and wherein the fourth module is configured to determine a sensitivity of the crop to environmental stresses based on the calculated crop injury.

[0016] The computer system may be configured as a distributed computer system. In particular, the method may be executed by different computing devices of the distributed computer system. In particular, the distributed computer system may comprise one or more servers and one or more client devices and the method may be executed in part by at least one of the servers and in part by at least one of the client devices. For example, the method may be carried out at least in part in cloud environment.

[0017] The system may comprise one or more devices other than the computer system. The system may comprise one or more imaging devices, e.g. one or more cameras, configured to acquire the aerial images, such as multispectral cameras. The system may comprise one or more vehicles to which the one or more imaging devices are mounted, such as airborne vehicles like drones.

[0018] The system, alternatively or in addition, may comprise a selection unit configured to select a crop to be planted and / or a crop to be used for breeding and / or a crop treatment on the basis of the determined sensitivity.

[0019] The present disclosure further provides a computer program product comprising instructions which, when carried out by a computer, cause the computer to carry out and / or control the method of the present disclosure.

[0020] The present disclosure further provides a computer-readable medium having stored thereon instructions which, when carried out by a computer, cause the computer to carry out and / or control the method of the present disclosure. The present disclosure further provides use of the determined sensitivity of a crop to environmental stresses for trait development of crops, breeding, crop protection, and / or crop selection.

[0021] In the present disclosure, the term “foliar damage component” is used interchangeably with the tern “discoloration component”.

[0022] As outlined above, disclosed is a computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising providing aerial image data based on one or more aerial images of a crop plot; calculating, based on aerial image data, a stunting component, a foliar damage component, and crop injury; calculating the crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury.

[0023] The term “environmental stresses”, as used herein, is to be understood broadly. They may also be referred to as environmental influences. Environmental stresses may comprise treatment with treatment product, particularly crop protection product. Treatment products may comprise herbicides, insecticides, fungicides, or the like. Environmental stress may comprise heat and / or drought and / or frost. These stresses are mere examples. The invention may entail other stresses. Moreover, stresses may refer to stress caused by a single source of stress or a combination of multiple sources stress.

[0024] The method described above may be carried out for one or more stresses, also referred to as sources of stress. The one or more stresses may be selected based on the application at hand. For example, where sensitivity to a treatment with an intended treatment product is to be determined, the selected stresses may be stresses caused by the intended treatment products. For example, the stress or stresses may be selected based on the aim, such as finding traits in a crop that leads to resistance to a particular stress.

[0025] The sensitivity of a crop to environmental stresses, as used herein, is to be understood broadly. The sensitivity need not be expressed by a specific number or in specific units. Sensitivity may, for example, be expressed by numerical values, by numerical ranges, by categories, or the like.

[0026] Sensitivity to stresses is to be understood as an indicator or representation for adverse effects on a crop, also referred to as damage to a crop, in response to the stresses. It is to be understood that the sensitivity of the crop may be different for different stresses. For example, stress due to heat and / or drought may differ from sensitivity due to treatment with a treatment product, such as herbicide, insecticide, or fungicide. Even within treatment products, a crop’s sensitivity may differ for different treatment products. As will be explained in detail below, adverse effects may include one or more of chloroses, necrosis, and stunting.

[0027] Providing aerial image data may comprise making aerial image data available for use in calculating a stunting component and a foliar damage component. Aerial image data to be provided may be aerial image data retrieved from a data storage. Aerial image data may be received directly from one or more imaging devices acquiring aerial images and / or may be received from one or more other devices.

[0028] The method may optionally comprise the step of acquiring aerial images prior to providing the aerial image data, e.g. by means of an imaging device. The imaging device may, for example, be mounted to an airborne vehicle like a drone. Aerial image data being based on aerial images is to be understood broadly and may entail that the aerial image data comprise the aerial images or a subset of the aerial images and / or that the aerial image data comprises data derived from the aerial images. For example, aerial images may be processed to obtain the aerial image data, such as converted, compressed, filtered, enhanced, or the like.

[0029] Aerial images of a crop plot are to be understood as images that depict at least part of, particularly the entire crop plot. In other words, the images may be acquired with a field of view of the imaging device comprising the at least part of the crop plot, particularly the entire crop plot. The images, in particular, may depict the at least part of the crop plot in a top view. An aerial image may depict multiple crop plots or parts of multiple crop plots.

[0030] In the following, the terms crop plot and crop will be explained in detail.

[0031] An agricultural field may comprise a piece of land that may be planted with a crop. A crop may comprise one or more crop plants. The agricultural field may be a test field, also referred to as experimental field. A test field may be a field for which at least some environmental parameters are controlled and / or monitored and where the crop plots may be subject to examination.

[0032] The agricultural field may be divided into plots of the agricultural field, wherein plots cover the entire agricultural field. These plots, herein, are referred to as crop plots so as to indicate that the plots are planted with a crop. The crop plots may all have the same shape and size. Alternatively, the crop plots may comprise portions of different shapes and / or of different sizes.

[0033] The method of the present disclosure may be carried out for the entire agricultural field or only for part of the agricultural field, for example, for one or more, particularly all, of the crop plots.

[0034] In addition to crop plants, harmful organisms may be present on a crop plot. Harmful organisms, according to the present disclosure, may comprise at least one of weeds, fungi, and pests. Pests may comprise insects.

[0035] As mentioned above, treatment of a crop with a treatment product may be an environmental stress. Treatment product may, for example, comprise crop protection product. Crop protection products may be products configured to prevent and / or mitigate growth of harmful organisms. Crop protection products may comprise at least one of fungicides, herbicides, and pesticides.

[0036] Crop protection products may comprise chemical products such as fungicides (e.g. of the chemical classes of the triazoles, the carboxamides, methoxyacrylates etc.), plant growth regulators (e.g. of the chemical classes of the triazoles, gibberilin inhibitors, quarternay ammonium salts etc.), insecticides (e.g. of the chemical class of the pyrethroides, juvenile hormone analoges etc.), herbicides, acaricides, molluscicides, nematicides, avicides, rodenticides, repellants, bactericides, biocides, safeners, urease inhibitors, nitrification inhibitors, denitrification inhibitors, foliar applied fertilizers (particular ammonia and / or nitrate based), other nutrients (e.g. micronutrients), seeds, seedlings, water or any combination thereof. Further, crop protection products may comprise biological products such as microorganisms useful as fungicide (e.g. Bacillus subtilis, Bacillus amyloliquefacients), herbicide (bioherbicide), insecticide (bioinsecticide e.g. Bacillus thuringiensis), or plant growth promoting bacteria (e.g. Bacillus amyloliquefacients, Bacillus pumilus etc.) acaricide (bioacaricide), molluscicide (biomolluscicide), nematicide (bionematicide- e.g. Bacillus firmus), avicide, piscicide, rodenticide, repellant, bactericide, biocide, safener, total herbicide / burndown herbicide (e.g. pelargonic acid), defoliant, desiccant, urease inhibitor, nitrification inhibitor, denitrification inhibitor, or any combination thereof. As explained above, based on the aerial image data, a stunting component and a foliar damage component are calculated. Crop injury is calculated through a weighted summation of the stunting component and the foliar damage component.

[0037] A stunting component is a parameter representative of stunting of the crop plants. Stunting is one of the detrimental or damaging effects that may be caused by environmental stresses. The stunting component may be a unitless parameter or have any suitable units. The stunting component may be a numerical value, for example. The stunting component, according to the present disclosure, may be understood as a stunting component for the entire crop plot. In other words, there may be exactly one value of the stunting component for one entire crop plot, particularly for exactly one crop plot. In yet other words, the stunting component may be plotspecific, and may also referred to as per-plot stunting component.

[0038] A foliar damage component is a parameter representative of discoloration of the crop plot, particularly the crop plot canopy. The discoloration may be caused by discoloration of the crop plants, for example their leaves. Discoloration of crop plants may be caused by detrimental or damaging effects caused by environmental stresses. The foliar damage component may be a unitless parameter or have any suitable units. The foliar damage component may be a numerical value, for example. The foliar damage component, according to the present disclosure, may be understood as a foliar damage component for the entire crop plot. In other words, there may be exactly one value of the foliar damage component for one entire crop plot, particularly for exactly one crop plot. In yet other words, the foliar damage component is plotspecific, and may also be referred to as per-plot foliar damage component.

[0039] Details concerning the stunting component and the foliar damage component are provided further below.

[0040] Crop injury may be a unitless parameter or have any suitable units. The crop injury may be a numerical value, for example. Crop injury may refer to an overall crop injury for the respective crop plot, e.g., not to the injury of individual crop plants. In other words, the crop injury, according to the present disclosure, may be understood as a crop injury for the entire crop plot. In other words, there may be exactly one value representative of the crop injury for one entire crop plot, particularly for exactly one crop plot. In yet other words, the crop injury is plot-specific, and may also referred to as per-plot crop injury.

[0041] As explained above, the crop injury is calculated through a weighted summation of the stunting component and the foliar damage component. The foliar damage component may be a component of at least one of: chlorosis, bleaching, and necrosis. In particular, the foliar damage component may be a combined component of at least two of: chlorosis, bleaching, and necrosis. In particular, the foliar damage component may be a combined chlorosis and necrosis component, This will be explained in more detail below.

[0042] For example, bleaching and chlorosis may each be seen as symptoms of reduced chlorophyl. Bleaching may be a loss of pigmentation in leaf tissues resulting in a whitish or pale appearance. Bleaching occurs when chlorophyll breaks down or is inhibited and is a symptom of environmental stress (including herbicide damage). Compared to chlorosis, bleaching is usually a more severe loss of chlorophyll. Necrosis may, for example, be seen as indicating tissue death. In the weighted summation, the crop injury, Cl, may equal a sum of the stunting component, SC, multiplied by a first weight w1 and the foliar damage component, FD, multiplied by a second weight w2.

[0043] In other words, the crop injury may be calculated using the following equation:

[0044] Cl = w1*SC + w2*FD, wherein Cl is the crop injury, SC is the stunting component, FD is the foliar damage component, w1 is a first weight, and w2 is a second weight.

[0045] The weights w1 and w2 may each be a positive number, particularly smaller than 1. There may be a functional relationship between w1 and w2. For example, w1 + w2 may equal 1. As will be explained below, w2 may be a function of SC or w1 a function of FD, for example.

[0046] As an example, w1 may be between 0.6 and 1, particularly 0.7 and 1, particularly 0.75 and 1. In particular, w1 may equal 1. The weight w2 may be a function of SC. As an example, w2 may be between 0.2 and 1, particularly 0.2 and 0.5. In particular, the values of said function may be within the range between 0.2 and 1 , particularly 0.2 and 0.5. The weights may, in particular be selected such that Cl does not exceed 100 % and / or such that the stunting components makes a higher contribution. This can be achieved by making w2 a function of w1 , for example.

[0047] As explained above, the method of the present disclosure comprises determining a sensitivity of the crop to environmental stresses based on the calculated crop injury. Moreover, as outlined above, the sensitivity of a crop to environmental stresses need not be expressed by a specific number or in specific units. Sensitivity may, for example, be expressed by numerical values, by numerical ranges, by categories, or the like.

[0048] The sensitivity may be crop specific. It may, particularly, not be plot-specific. That is, the sensitivity may result from the crop injuries of multiple crop plots, for example based on statistical calculations. Determining the sensitivity may comprise a calculation. The calculation may take the calculated crop injury of one crop plot or of multiple crop plots, particularly all crop plots, as an input and yield a value that is the sensitivity of the crop as an output or yield a value that is associated with a sensitivity of the crop as an output. A value may be associated with the sensitivity of the crop, for example, by a mapping and / or based on rules and / or based on a table. For example, a value may be in one of a plurality of ranges, and each range may correspond to a sensitivity. Optionally, the calculation may entail carrying out statistical computations, particularly where the input comprises the crop injury of multiple crop plots.

[0049] Alternatively the determining may take the calculated crop injury of one crop plot or of multiple crop plots, particularly all crop plots as an input and yield the sensitivity of the crop as an output without carrying out calculations, for example, by applying rules defining the relation between input and output values.

[0050] Thus, to summarize, the present disclosure provides a method that allows for determining the sensitivity of a crop plot to environmental stresses. This can be done in an objective way by analyzing aerial image data to quantitatively derive crop injury. The proposed method is less error prone, more reproducible, and, accordingly, more reliable than presently known methods for estimating sensitivity of a crop.

[0051] Therefore, it can also be used as an objective basis for selecting a crop, developing traits, and / or breeding. It is to be understood that, for example, multiple crop plots may be exposed to one or more environmental stresses on purpose in a controlled environment, such as a trial field. The controlled environment may also have one or more control crop plots not exposed to the selected environmental stresses, e.g. for use as references. This will be explained in detail further below.

[0052] According to the present disclosure, the method may comprise extracting per-plot statistics and deriving the stunting component and / or the foliar damage component from the per-plot statistics. The per-plot statistics may be derived from the aerial image data.

[0053] The method may comprise deriving per-plot statistics from the aerial image data, e.g. extracting the per-plot statistics from the aerial image data or from data derived therefrom. The stunting component and / or the foliar damage component may be derived from said per-plot statistics.

[0054] Determining per-plot statistics may comprise counting of numbers of vegetation pixels per-plot and / or calculating a mean value and / or on another statistical value characterizing the plot, such as a vegetation index of the plot. More details are provided further below.

[0055] As an example, a multispectral (e.g. 5-band) reflectance image may be converted into a singleband (grayscale) MCARI2 image, where every pixel has an MCARI2 value. Per-plot means may be calculated by averaging MCARI2 values of all vegetation pixels of a plot, e.g. all vegetation pixels delineated by a plot polygon.

[0056] Vegetation pixels may be pixels that are automatically determined or categorized as depicting vegetation. For example, vegetation pixels may be distinguished from soil pixels. The determining or categorizing may, for example, be based on a model or rules. As an example, color of a pixel or context of the pixel in the image may be used for the determining or categorizing. As an example, a value of a vegetation index of the pixel may be used to categorize this pixel. More details are provided further below.

[0057] Using statistical methods for investigating the crop plots, particularly determining the components using statistical methods, allows for a better overall result and increased comparability.

[0058] According to the present disclosure, the crop plot may be part of a trial field comprising a plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots. In this case, the method may comprise extracting trial field per-plot statistics comprising per-plot statistics for the treated crop plots and per-plot statistics for the untreated control crop plots, and calculating the stunting component and the foliar damage component may comprise deriving the stunting component and / or the foliar damage component from the trial field per-plot statistics.

[0059] Deriving the stunting component and / or the foliar damage component from the trial field per-plot statistics may be based on a comparison of the per-plot statistics for the treated crop plots and the per-plot statistics for the untreated control crop plots.

[0060] The trial field per-plot statistics may comprise a first set and a second set of trial field per-plot statistics, the first and second sets being calculated using different statistical metrics. In this case, calculating the stunting component and the foliar damage component may comprise deriving the stunting component from the first set of trial field per-plot statistics and deriving the foliar damage component from the second set of trial field per-plot statistics. In other words, calculating the stunting component and the foliar damage component may comprise deriving the stunting component from trial field per-plot statistics having been calculated using different statistical metrics than the foliar damage component.

[0061] In yet other words, calculating the stunting component and the foliar damage component may comprise deriving the stunting component and the foliar damage component from trial field per plot statistics having been calculated using different statistical metrics for the stunting component and the foliar damage component.

[0062] According to the present disclosure, the the per-plot statistics may be extracted from vegetation- only image data derived from the aerial image data.

[0063] Vegetation-only image data may be derived from the aerial image data by removing or maskingout non-vegetation pixels, such as soil pixels, from the aerial image data. Non-vegetation pixels may be pixels that are automatically determined or categorized as not depicting vegetation. Soil pixels may be pixels that are automatically categorized as depicting soil. The determining or categorizing may, for example, be based on a model or rules. As an example, color of a pixel or context of the pixel in the image may be used for the determining or categorizing.

[0064] As an example, Optimized Soil Adjusted Vegetation Index (OSAVI) may be used to classify pixels as being vegetation pixels or non-vegetation pixels, i.e. , into vegetation and non- vegetation classes.

[0065] Using vegetation-only image data allows for improved quality of results. That is, since it is a goal to study how stresses affect the crop, any pixels that do not depict vegetation would introduce noise in the results.

[0066] According to the present disclosure, the method may comprise generating scouting maps with respect to vegetation indices based on the vegetation-only image data and deriving the per-plot statistics from the scouting maps.

[0067] A scouting map with respect to a vegetation index may also be referred to as a vegetation index scouting map. The scouting map may be a map that provides georeferenced vegetation index data. Such scouting maps may be obtained by collecting image data, and processing the collected image data including removing non-vegetation pixels, such as soil pixels, and generating vegetation-only image data that are georeferenced. Based thereon, the vegetation indices may be determined and vegetation index scouting maps, such as MCARI2, NDRE, or NDVI scouting maps, may be created.

[0068] The use of a scouting map for deriving the per-plot statistics is a particularly effective means, as the georeferencing allows for easily correlating vegetation indices with a corresponding crop plot.

[0069] According to the present disclosure the per-plot statistics, particularly the second set of trial field per-plot statistics, may comprise vegetation indices. Alternatively or in addition, the per-plot statistics, particularly the first set of trial field per-plot statistics, may comprise a number of vegetation pixels. In particular, the number of vegetation pixels may be counted based on vegetation-only image data derived from the aerial image data.

[0070] As already described above, the stunting component and / or the foliar damage component may be derived from per-plot statistics. The per plot statistics may be based on determining numbers of pixels in aerial image data for a given plot. For example, the number of vegetation pixels of a given plot may be determined from the aerial image data or data derived therefrom, as described above. In this case, in particular, the above-described vegetation-only image data may be used for determining the number of vegetation pixels. Alternatively or in addition, the stunting component and / or the foliar damage component may be derived from one or more vegetation indices. Such indices, in principle, are known in the art for other purposes.

[0071] The above-described per-plot statistics allow for improved quality of results, particularly improved comparability.

[0072] According to the present disclosure, the vegetation indices may comprise Modified Chlorophyll Absorption Ratio Index Improved, MCARI2, and / or Normalized Difference Red Edge Index, NDRE, and / or Normalized Difference Vegetation Index, NDVI.

[0073] MCARI2 is considered a predictor of green leaf area. It indicates relative abundance of chlorophyll.

[0074] NDRE is a vegetation index that can be used to analyze multi-spectral image data to determine whether vegetation therein is healthy. NDRE is based on a ratio of spectral reflectance of nearinfrared and edge of red derived from multispectral images.

[0075] NDVI is an index that quantifies vegetation greenness. For example, it may be useful in understanding vegetation density and plant health. NDVI is based on a ratio of spectral reflectance of red and near-infrared derived from multispectral images.

[0076] An advantage of using these or similar indices is that they have good predictive quality for chlorosis, necrosis, and stunting, and accordingly are useful for determining the stunting component and the foliar damage component.

[0077] As an example, the foliar damage component may correspond to or be a function of one of said vegetation indices, optional a function of a relative vegetation index, where the relative vegetation index is calculated from the vegetation index the crop plot and one or more control crop plots.

[0078] According to the present disclosure, the method may comprise, prior to extracting the per-plot statistics, pre-processing aerial image data, particularly to obtain the vegetation-only image data. The pre-processing may comprise at least one of georeferencing, orthomosaic generation, and soil pixel removing.

[0079] Georeferencing as part of pre-processing the aerial image data allows for easily correlating the pre-processed data with corresponding crop plots and thereby for effective determination of per- plot statistics.

[0080] Orthomosaic generation as part of pre-processing allows for particularly efficient and precise determination of per-plot statistics, as it allows for combining multiple images and correcting geometric distortion. Any known method for orthomosaic generation may be used.

[0081] Soil pixel removal as part of the pre-processing allows for arriving at vegetation-only image data in an efficient manner.

[0082] According to the present disclosure, the foliar damage component may be a component of at least one of: chlorosis, bleaching, and necrosis. In particular, the foliar damage component may be combined component of at least two of chlorosis, bleaching, and necrosis, in particular a combined component of chlorosis and necrosis. In other words, rather than identifying chlorosis, bleaching, and necrosis separately, a component, referred to as foliar damage component, is identified. Chlorosis, bleaching, and necrosis can be caused by environmental stresses. Accordingly, they are good indicators for stress sensitivity of a crop. Chlorosis and bleaching, i.e. a lack of chlorophyll, and necrosis, i.e., cell death, represent foliar damage, and in particular, may each lead to discoloration. Combining them into a combined foliar damage component is efficient and yields accurate results. As explained above, such a foliar damage component may correspond to or be a function of a vegetation index.

[0083] According to the present disclosure, a weight of the foliar damage component FD may be a, particularly non-linear, function of the stunting component SC.

[0084] For example, a weighted sum wherein the weight of the foliar damage component FD is a nonlinear, function of the stunting component SC may look as follows:

[0085] Cl = SC + (a*SC2+ b*SC + c)*FD, wherein a, b, and c, are constants of real numbers. A non-limiting example for a calculation of the weighted sum with nonlinear function is shown below.

[0086] Cl = SC + (1 - 0.0001 *SC2+0.002*SC)*FD

[0087] An advantage of making the weight of the foliar damage component a function of the stunting component is that it allows to ensure that stunting is the more important contribution to the overall crop injury value. That is, when the stunting component is high, the contribution of the foliar damage goes down. This also allows overall crop injury to not exceed 100%. An advantage of this is that it is taken into account that stunting usually has the stronger effect on yield. Alternatively, the weight of the stunting component may be a function of the foliar damage component.

[0088] According to the present disclosure, the crop plot may be part of a / the trial field comprising a / the plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots.

[0089] Treatment of a crop with a treatment product may be considered to be an environmental stress. The treatment product may, for example, be a herbicide, fungicide, pesticide, or a combination of two or more of these, as described above. Thus, it can be studied how the treatment product affects the crop, or in other words, how sensitive the crop is to the treatment with the treatment product.

[0090] The trial field per-plot statistics may comprise a / the number of vegetation pixels and the stunting component may be calculated based on a comparison of the respective number of vegetation pixels for a treated crop plot and an untreated control plot of the trial field. Such a comparison is is an efficient way of objectively quantifying stunting.

[0091] Alternatively or in addition, the trial field per-plot statistics may comprise vegetation indices and the foliar damage component may be calculated based on a comparison of the respective vegetation indices of a treated crop plot and an untreated control crop plot of the trial field.

[0092] The above can be applied in the same manner for other types of environmental stress.

[0093] Thus, the method of the present disclosure allows for determining the effect of an environmental stress relative to a reference, i.e., a crop that is not exposed to the stress and / or resistant to the stress. Results of this allow for better distinguishing the effects of the specific environmental stress from other effects that are not associated with this specific environmental stress. Accordingly, accuracy can be improved.

[0094] According to the present disclosure, the crop plot may be part of a / the trial field comprising a / the plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots. The calculating of the stunting component may comprise calculating, in particular based on respective numbers of vegetation pixels for a treated crop plot and an untreated control crop plot, a relative vegetation cover of the treated crop plot. A relative vegetation cover is an efficient way of objectively quantifying stunting.

[0095] Alternatively or in addition, calculating of the foliar damage component may comprise calculating, in particular based on respective vegetation indices for a treated crop plot and an untreated control crop plot, a relative vegetation index of at least one of, in particular at least two of, in particular all of: chlorosis, bleaching, and necrosis. In particular, calculating of the foliar damage component may comprise calculating, in particular based on respective vegetation indices for a treated crop plot and an untreated control crop plot, a relative vegetation index of chlorosis and necrosis.

[0096] This can be applied in the same manner for other types of environmental stress.

[0097] Thus, the method of the present disclosure allows for determining the effect of an environmental stress relative to a reference, i.e. , a crop that is not exposed to the stress. Results of this allow for better distinguishing the effects of the specific environmental stress from other effects that are not associated with this particular environmental stress. Accordingly, accuracy can be improved.

[0098] According to the present disclosure, a virtual machine may be used for one or more, in particular all, steps described above, particularly a Geographic Information System, GIS, virtual machine.

[0099] When using a GIS virtual machine, the stunting component, the foliar damage component, and / or the overall crop injury may be stored in the virtual machine, for example as (new) attributes to corresponding shapefiles in the GIS virtual machine, particularly for further processing and / or storage and / or analysis.

[0100] The method of the present disclosure may comprise exposing one or more crop plots of a / the trial field to the environmental stress, such as treating the one or more crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, and determining the sensitivity of the crop on the basis of the crop injury for at least some, in particular all, of the exposed crop plots.

[0101] The method according to the present disclosure may comprise exposing only a subset of all crop plots of a / the trial field to the environmental stress, such as treating the subset of crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, determining the sensitivity on the basis of the crop injury for at least some, in particular all, of the exposed crop plots, wherein determining the crop injury and / or determining the sensitivity is based taking into account aerial image data of the unexposed crop plots, for example by using the unexposed crop plots as a reference.

[0102] The method according to the present disclosure may comprise at least one of: selecting, among a plurality of candidate crops, a crop to be planted on the basis of the determined sensitivity, and / or selecting, among a plurality of candidate crops, a crop for breeding on the basis of the determined sensitivity.

[0103] In other words, the determined sensitivity may be used as a decision criterion. The sensitivity may be used, for example, to be determine which crop to plant on a piece of land, e.g., so as to select a crop that is particularly resistant to specific stresses. The sensitivity may, alternatively or in addition, be used to select crops to be used for breeding, e.g., so as to enhance this property of a crop.

[0104] BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In the following, the present disclosure is further described with reference to the enclosed figures:

[0106] Fig. 1 illustrates an exemplary system in which the method of the present disclosure may be carried out.

[0107] Fig. 2 illustrates an exemplary method according to the present disclosure.

[0108] Fig. 3 illustrates a method for calculating the stunting component for an experimental plot according to the present disclosure.

[0109] Figs. 4a and 4b illustrate a process of deriving a vegetation-only image according to the present disclosure.

[0110] Fig. 5 illustrates an exemplary scatter plot of visually assessed crop injury and crop injury derived from images.

[0111] DETAILED DESCRIPTION OF EMBODIMENTS

[0112] The following embodiments are mere examples for implementing the invention disclosed herein and shall not be considered limiting.

[0113] Figure 1 shows a system 1 according to the present disclosure. The system comprises a computing system 2 configured to carry out a method for determining a sensitivity of a crop to environmental stresses, the method comprising providing aerial image data based on one or more aerial images of a crop plot; calculating, based on aerial image data, a stunting component and a foliar damage component; calculating crop injury through a weighted summation of the stunting component and the foliar damage component; and determining a sensitivity of the crop to environmental stresses based on the calculated crop injury. The computing system may be a centralized or a distributed computing system.

[0114] The system optionally comprises one or more imaging devices 3 for capturing the aerial images. The one or more imaging devices are shown in Figure 1 as mounted on a drone 4 and a satellite 5 as non-limiting examples. However, the imaging devices may be mounted, for example, to any kind of airborne vehicle. The imaging devices may, for example, be configured to provide multispectral images of the crop plot. As an example, the imaging devices may comprise a camera that may, for example, be mounted to a drone.

[0115] Figure 1 , for the sake of illustration, also shows a data storage device 6 that is external to the imaging devices and the computing system. The data storage device may be configured to exchange data with the imaging devices and the imaging systems, for example by wired or wireless data connections 7. The system may be configured such that aerial image data is provided by the imaging devices to the storage device and provided by the storage device to the computing system. Alternatively or in addition, the system may be configured such that the imaging devices provide the aerial image data directly to the computing system via wired or wireless data connections 8.

[0116] For the sake of illustration, Figure 1 also shows an agricultural field 10, divided into crop plots 11. As explained above, the agricultural field may have treated crop plots and untreated crop plots. In Figure 11, treated crop plots are indicated by reference sign 11a and untreated crop plots by reference sign 11b. However, it is noted that the method may also be carried out without untreated crop plots.

[0117] Optionally, the system, particularly the computing system 2, may comprise a selection unit 2a configured to select a crop to be planted and / or a crop to be used for breeding on the basis of the determined sensitivity.

[0118] Figure 2 illustrate an exemplary method according to the present disclosure.

[0119] In step S12, aerial image data based on one or more aerial images of a crop plot are provided.

[0120] In step S13, a stunting component and a foliar damage component are calculated based on the aerial image data. The foliar damage component may be a component of at least one of chlorosis, bleaching, and necrosis, in particular a combined component of chlorosis, necrosis and / or bleaching, particularly at least chlorosis and necrosis. Vegetation indices may be used for calculation of the components. The vegetation indices may comprise Modified Chlorophyll Absorption Ratio Index Improved, MCARI2, and / or normalized difference red edge index, NDRE, and / or normalized difference vegetation index, NDVI. In particular, the foliar damage component may be a function of one of the vegetation indices. Stunting and foliar damage components, in one example, are calculated relative to a control plot. An example for this is described further below.

[0121] In step S14, crop injury is calculated through a weighted summation of the stunting component and the foliar damage component. The weight of the foliar damage component may optionally be a, particularly non-linear, function of the stunting component or vice versa. The crop injury may be plot-specific, i.e. , determined individually for each crop plot.

[0122] In step S15, sensitivity of the crop to environmental stresses is determined based on the calculated crop injury. The sensitivity may be an overall sensitivity determined on the basis of the calculated crop injury of multiple, particularly all, of the crop plots.

[0123] In optional step S16, the crop injury and / or the sensitivity is output to a user.

[0124] In optional step S17, crop injury and / or sensitivity are used to select a crop to be planted on another agricultural field.

[0125] In optional step S18, a crop is selected for breeding based on crop sensitivity.

[0126] In optional step S11, which precedes claim S12, aerial images of the crop plot are acquired by an imaging device.

[0127] In optional step S10, which precedes claim S11, one or more crop plots are exposed to the environmental stress, for example treated by a treatment product. As an example, only a subset of crop plots of an agricultural field is intentionally exposed to the environmental stress in step S10a, whereas the remaining crop plots are not intentionally exposed to the environmental stress. Optionally, in step S10b, the remaining crop plots are protected from exposure to the environmental stress.

[0128] The method of the present disclosure, particularly as described above, may in particular be carried out by a system according to the present disclosure, for example as illustrated in and described in the context of Figure 1.

[0129] In the following further advantages and features of the method according to the present disclosure will be provided.

[0130] Crop injury, particularly quantification thereof, is useful for determining tolerance of a crop to certain stresses. For example, it may be used for determining herbicide tolerance of a crop. This may be done by means of aerial images, particularly multispectral images of crop plots.

[0131] One use case of the present disclosure is, thus, quantification of crop injury in herbicide tolerance trials using aerial multispectral imaging.

[0132] The method of the present disclosure allows for replacing subjective visual assessment of crop injury by a person or persons with an objective assessment, that can be carried out automatically and at high throughput.

[0133] According to the present disclosure, in an exemplary method, aerial multispectral images of field trials are collected when crop injury assessment needs to be performed. Per-plot mean parameters characterizing canopy vigour and size can be extracted from field images where non-vegetation pixels are removed or masked out.

[0134] Relative crop injury components such as stunting, and combined chlorosis and necrosis are calculated using the extracted per-plot parameters and information about treatments. In an example, this may be done in such way that untreated control plots have stunting, chlorosis and necrosis values of 0. Per-plot relative crop injury components are combined as a weighted sum, where weight of stunting is 1 and weight of chlorosis and necrosis is a function of stunting, into an overall crop injury. The information about treatments comprises information defining which plots or areas of the field should be used as zero-crop-injury reference, e.g., untreated control plots in case of treatment product, such as herbicide, application. Reference is made to the equations above as an example how the components may be calculated.

[0135] Compared to manual visual assessment of crop injury, the image-based method has a higher throughput and generates objective (reproducible) data, thereby being more reliable.

[0136] In other words, an instrument-based high-throughput method for assessment of crop injury can be provided, e.g., for soy herbicide tolerance trials.

[0137] The method of the present disclosure can be applied for crop injury quantification for different types of crops, e.g., soybeans, cotton, canola, etc., without the need for development of a cropspecific prediction model.

[0138] The method of the present disclosure may be employed for quantification of stresses which result in canopy chlorosis and necrosis or stunting in field experiments where stress-free plots are available. For example, quantification of Iron Deficiency Chlorosis (I DC) in soybean trials containing I DC-resistant checks may be carried out.

[0139] The present disclosure can process aerial images, such as multispectral bands of an aerial image to determine canopy area, height, volume, greenness, senescence, and / or vegetation indices. In an optional example, at least one additional image, specifically an initial image, e.g. aerial image, of the field may be collected right before exposing the crop to an environmental stress, e.g. right before application of a herbicide. Initial crop injury values may be calculated from the initial image, for example in the manner described herein. These initial crop injury values, e.g. before application of herbicide, may be related to any field variations (e.g. as a background effect) and may be subtracted from all future calculated crop injury values after exposure to environmental stress, e.g. after application of the herbicide.

[0140] Although any suitable hardware and software tools may be used in the method and system of the present disclosure, exemplary hardware may comprise one or more imaging devices like multispectral or hyperspectral imaging equipment for acquiring the aerial image data and / or an aerial vehicle, such as a drone, onto which (the) imaging device(s) are mounted during acquisition of the aerial image data by the imaging device(s).

[0141] An exemplary calculation that may be employed in the method of the present disclosure is shown below.

[0142] A stunting component SC is (directly) calculated as a relative vegetation cover, for example as follows:

[0143] SC = (1 - VCExp / VCUTc)*100o / o where VCEXPand VCUTCare fractions of vegetation cover in experimental plots and untreated control plots, respectively. Experimental plots are those exposed to the stress, e.g. to the treatment. The fraction of vegetation cover may, for example, be calculated as a ratio of number of vegetation pixels to the number of all pixels of a plot, e.g. a ratio of number of vegetation pixels to the number of all pixels in a plot polygon.

[0144] A combined chlorosis and necrosis component FD is calculated directly as a relative vegetation index, for example as follows:

[0145] FD = (1 - VIEXP / VIUTC)*100O / O where VIEXp and VIUTCmay, for example be values of MCARI2 or NDRE vegetation indices in experimental and untreated control plots, respectively.

[0146] The index may be selected according to the scenario at hand. For example, MCARI2 may be suitable where a wider range of chlorosis and necrosis and / or multiple MOAs are expected and NDRE where lower range of chlorosis and necrosis and / or single MOA is expected.

[0147] In this context, MOA refers to mode of action. A treatment product, such as herbicide, may have one or moder MOAs. An MOA may describe how the treatment product affects vegetation.

[0148] Different MOAs, for example, may define symptoms (chlorosis, necrosis, bleaching, leaf curling, ...) that are derivable and quantifiable using aerial image data, such as multispectral image data. They may then be expressed by an overall crop injury value.

[0149] The crop injury, referred to Cl, may be calculated by a weighted sum of SC and FD.

[0150] The weighted sum of SC and FD may be calculated as follows:

[0151] Cl = 0.75*SC + 0.25*FD

[0152] Alternatively, a weighted sum may be calculated as follows:

[0153] Weighted sum of SC and FD Cl = SC + (1 - 0.000 SC2+0.002*SC)*FD

[0154] Here, the weight given to the foliar damage component FD is a non-linear function of the stunting component.

[0155] The calculation may be preceded by collecting and processing raw multispectral aerial images at each time point when crop injury needs to be assessed.

[0156] Soil pixels may be removed from processed multispectral images, vegetation-only MCARI2 and NDRE indices may be calculated, and per-plot means may be extracted.

[0157] The per-plot means and untreated control plot information may be used for calculating stunting, and combined chlorosis necrosis, and overall crop injury.

[0158] In case the method should be validated, visually assessed crop injury may additionally be used (e.g. as ground truth).

[0159] Thus, to summarize, the method of the present disclosure allows for assessing crop injury in field trials, for example during development of new herbicides or herbicide resistance traits in crops. Currently, crop injury is estimated subjectively, by visual assessment, which is timeconsuming and subjective.

[0160] The method of the present disclosure may be used in trait development of crops, breeding, and / or crop selection.

[0161] The method is time saving and has objective and reproducible results. It is applicable for different types of crops. It is also less time-consuming than potential machine learning approaches, which would require collection of training data first, potentially over multiple seasons.

[0162] Fig. 3 illustrates a method for calculating the stunting component for an experimental plot according to the present disclosure.

[0163] First, a number of total pixels and vegetation pixels per plot is determined. Then, a fraction of vegetation coverage in the experimental plot is determined, e.g., \ / CEXp = 0.1.

[0164] In addition, a Fraction of vegetation coverage in untreated control plot is determined, e.g. \ / UTC = 0.4.

[0165] Next, the stunting of the experimental plot is determined as a relative vegetation coverage, e.g., SC = (1 - 0.1 / 0.4)*100 % = 75 %. It is noted that the stunting of an untreated control plot is 0 % by definition.

[0166] The same principle may be used for calculation of a foliar damage component FD. Instead of fraction of vegetation coverage, mean values of vegetation index, for example MCARI2 or NDRE, of experimental plot and untreated control plots can be used.

[0167] Figs. 4a and 4b illustrate, in different representations, a process of deriving a vegetation-only image according to the present disclosure. In this example, for illustrative purposes, Optimized Soil Adjusted Vegetation Index (OSAVI) was used to identify a threshold for classification of pixels into vegetation and non-vegetation classes. The Figure shows, from left to right, an initial multispectral aerial image, an OSAVI image with pixel values for identification of appropriate threshold, and a mask based on OSAVI thresholding used for removal of non-vegetation pixels. Fig. 5 illustrates an exemplary scatter plot of visually assessed crop injury and crop injury derived from images using derived from a soybean field trial used for testing performance of the method, e.g., to see how image-based crop injury values align with visually assessed crop injury values. The method of the present disclosure was used to obtain the derived crop injury. In this example, each data point represents a treatment average calculated from 4 plot replicates and assessed at 3 time points. In other words, each point in this particular scatter plot represents an average visually assessed crop injury (x-axis) and image-based crop injury (y-axis) averaged across replicated field plots (4 replicates, for example) that received the same treatment and were assessed 3 times (1 week after treatment, 2 weeks after treatment, and 3 weeks after treatment, for example).

[0168] EXAMPLES

[0169] 1. A computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising: providing (S12) aerial image data based on one or more aerial images of a crop plot, calculating (S13), based on the aerial image data, a stunting component and a discoloration component; calculating (S14) crop injury through a weighted summation of the stunting component and the discoloration component; determining (S15) a sensitivity of the crop to environmental stresses based on the calculated crop injury.

[0170] 2. The method according to Example 1, comprising extracting per-plot statistics and deriving the stunting component and / or the discoloration component from the per-plot statistics that are derived from the aerial image data.

[0171] 3. The method according to Example 2, wherein the per-plot statistics are extracted from vegetation-only image data derived from the aerial image data.

[0172] 4. The method according to Example 2 or 3, comprising generating scouting maps with respect to vegetation indices based on the vegetation-only image data and deriving the per-plot statistics from the scouting maps.

[0173] 5. The method according to any of Examples 2 to 4, wherein the per-plot statistics comprise vegetation indices and / or a number of vegetation pixels, in particular wherein the number of vegetation pixels is counted based on vegetation-only image data derived from the aerial image data.

[0174] 6. The method according to any of Examples 4 to 5, wherein the vegetation indices comprise Modified Chlorophyll Absorption Ratio Index Improved, MCARI2, and / or normalized difference red edge index, NDRE, and / or normalized difference vegetation index, NDVI.

[0175] 7. The method according to any of Examples 3 to 6, the method comprising, prior to extracting the per-plot statistics, pre-processing aerial image data, wherein the preprocessing comprises at least one of: georeferencing, orthomosaic generation, and soil pixel removing.

[0176] 8. The method according to any of the preceding Examples, wherein the discoloration component is a combined component of chlorosis and necrosis. The method according to any of the preceding Examples, wherein a weight of the discoloration component is a, particularly non-linear, function of the stunting component. The method according to any of Examples 2 to 9, wherein the crop plot is part of a trial field comprising a plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots, wherein the trial field per-plot statistics comprise a / the number of vegetation pixels and wherein the stunting component is calculated based on a comparison of the respective number of vegetation pixels for a treated crop plot and an untreated control plot of the trial field, and / or wherein the trial field per-plot statistics comprise vegetation indices and wherein the discoloration component is calculated based on a comparison of the respective vegetation indices of a treated crop plot and an untreated control crop plot of the trial field. The method according to Example 2 to 9, wherein the crop plot is part of a trial field comprising a plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots, wherein the calculating of the stunting component comprises calculating, in particular based on respective numbers of vegetation pixels for a treated crop plot and an untreated control crop plot, a relative vegetation cover of the treated crop plot, and / or wherein calculating of the discoloration component comprises calculating, in particular based on respective vegetation indices for a treated crop plot and an untreated control crop plot, a relative vegetation index of chlorosis and necrosis. The method according to any of the preceding Examples, wherein a virtual machine is used for one or more, in particular all, steps of the preceding Examples, and / or wherein the stunting component, the discoloration component, and / or the overall crop injury are stored in a / the virtual machine, particularly for further processing and / or storage and / or analysis, in particular stored as new attributes to corresponding shapefiles in a Geographic Information System, GIS, virtual machine. The method according to any of the preceding Examples, further comprising exposing one or more crop plots of a / the trial field to the environmental stress, such as treating the one or more crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, determining the sensitivity on the basis of the crop injury for at least some, in particular all, of the exposed crop plots. The method according to any of the preceding Examples, further comprising exposing only a subset of all crop plots of a / the trial field to the environmental stress, such as treating the subset of crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, determining the sensitivity on the basis of the crop injury for at least some, in particular all, of the exposed crop plots, wherein determining the crop injury and / or determining the sensitivity is based taking into account aerial image data of the unexposed crop plots, for example by using the unexposed crop plots as a reference. The method of any of the preceding Examples, comprising at least one of: selecting, among a plurality of candidate crops, a crop to be planted on the basis of the determined sensitivity, and / or selecting, among a plurality of candidate crops, a crop for breeding on the basis of the determined sensitivity.

[0177] 16. A system (1) comprising a computing system (2) configured to carry out the method of any of the preceding Examples.

[0178] 17. The system of Example 16, comprising a selection unit (2a) configured to select a crop to be planted and / or a crop to be used for breeding on the basis of the determined sensitivity.

[0179] 18. A computer program product comprising instructions which, when the program is executed by a computing system, cause the computing system to carry out the method of any of Examples 1 to 15.

[0180] 19. A computer-readable medium having stored thereon instructions which, when the program is executed by a computing system, cause the computing system to carry out the method of any of Examples 1 to 15.

[0181] 20. Use of the determined sensitivity of a crop to environmental stresses determined with the method of any one of Examples 1 to 15 for trait development of crops, breeding, crop protection, and / or crop selection.

[0182] The present disclosure has been described in conjunction with preferred embodiments and examples as well. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed invention, from the studies of the drawings, this disclosure, and the claims. Notably, unless otherwise specified, any steps presented can be performed in any order, i.e., the present invention is not limited to a specific order of these steps. Moreover, unless specified otherwise, it is also not required that the different steps are performed at a certain place or at one computing device of a distributed system, i.e., each of the steps may be performed at different computing devices.

Claims

CLAIMS1. A computer-implemented method for determining sensitivity of a crop to environmental stresses, comprising: providing (S12) aerial image data based on one or more aerial images of a crop plot, calculating (S13), based on the aerial image data, a stunting component and a foliar damage component; calculating (S14) crop injury through a weighted summation of the stunting component and the foliar damage component; determining (S15) a sensitivity of the crop to environmental stresses based on the calculated crop injury.

2. The method according to claim 1, comprising extracting per-plot statistics and deriving the stunting component and / or the foliar damage component from the per-plot statistics that are derived from the aerial image data.

3. The method according to claim 1, wherein the crop plot is part of a trial field comprising a plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots, wherein the method comprises extracting trial field per-plot statistics comprising per-plot statistics for the treated crop plots and per-plot statistics for the untreated control crop plots, and wherein calculating the stunting component and the foliar damage component comprises deriving the stunting component and / or the foliar damage component from the trial field per-plot statistics.

4. The method according to claim 3, wherein deriving the stunting component and / or the foliar damage component from the trial field per-plot statistics is based on a comparison of the per-plot statistics for the treated crop plots and the per-plot statistics for the untreated control crop plots.

5. The method according to claim 3 or 4, wherein the trial field per-plot statistics comprise a first set and a second set of trial field per-plot statistics, the first and second sets being calculated using different statistical metrics, and wherein calculating the stunting component and the foliar damage component comprises deriving the stunting component from the first set of trial field per-plot statistics and deriving the foliar damage component from the second set of trial field per-plot statistics.

6. The method according to any of claims 2 to 5, wherein the per-plot statistics are extracted from vegetation-only image data derived from the aerial image data.

7. The method according to any of claims 2 to 6, comprising generating scouting maps with respect to vegetation indices based on the vegetation-only image data and deriving the per-plot statistics from the scouting maps.

8. The method according to any of claims 2 to 7, wherein the per-plot statistics, particularly the second set of trial field per-plot statistics, comprise vegetation indices, and / or wherein the per-plot statistics, particularly the first set of trial field per-plot statistics, comprise a number of vegetation pixels, in particular wherein the number of vegetationpixels is counted based on vegetation-only image data derived from the aerial image data.

9. The method according to any of claims 7 to 8, wherein the vegetation indices comprise Modified Chlorophyll Absorption Ratio Index Improved, MCARI2, and / or normalized difference red edge index, NDRE, and / or normalized difference vegetation index, NDVI.

10. The method according to any of claims 2 to 9, the method comprising, prior to extracting the per-plot statistics, pre-processing aerial image data, wherein the pre-processing comprises at least one of: georeferencing, orthomosaic generation, and soil pixel removing.

11. The method according to any of the preceding claims, wherein the foliar damage component is a component of at least one of: chlorosis, bleaching, and necrosis, in particular a combined component of at least two of: chlorosis, bleaching, and necrosis, particularly a combined component of chlorosis and necrosis.

12. The method according to any of the preceding claims, wherein a weight of the foliar damage component is a, particularly non-linear, function of the stunting component.

13. The method according to any of claims 2 to 12, wherein the crop plot is part of a / the trial field comprising a / the plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots, wherein the trial field per-plot statistics comprise a / the number of vegetation pixels and wherein the stunting component is calculated based on a comparison of the respective number of vegetation pixels for a treated crop plot and an untreated control plot of the trial field, and / or wherein the trial field per-plot statistics comprise vegetation indices and wherein the foliar damage component is calculated based on a comparison of the respective vegetation indices of a treated crop plot and an untreated control crop plot of the trial field.

14. The method according to claim 2 to 12, wherein the crop plot is part of a / the trial field comprising a / the plurality of crop plots, the plurality of crop plots comprising treated crop plots, treated with a treatment product, and untreated control crop plots, wherein the calculating of the stunting component comprises calculating, in particular based on respective numbers of vegetation pixels for a treated crop plot and an untreated control crop plot, a relative vegetation cover of the treated crop plot, and / or wherein calculating of the foliar damage component comprises calculating, in particular based on respective vegetation indices for a treated crop plot and an untreated control crop plot, a relative vegetation index of at least one of, in particular at least two of, in particular all of: chlorosis, bleaching, and necrosis, particularly a relative vegetation index of chlorosis and necrosis.

15. The method according to any of the preceding claims, wherein a virtual machine is used for one or more, in particular all, steps of the preceding claims, and / or wherein the stunting component, the foliar damage component, and / or the overall crop injury are stored in a / the virtual machine, particularly for further processing and / or storage and / or analysis, in particular stored as new attributes to corresponding shapefiles in a Geographic Information System, GIS, virtual machine.

16. The method according to any of the preceding claims, further comprisingexposing one or more crop plots of a / the trial field to the environmental stress, such as treating the one or more crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, determining the sensitivity on the basis of the crop injury for at least some, in particular all, of the exposed crop plots.

17. The method according to any of the preceding claims, further comprising exposing only a subset of all crop plots of a / the trial field to the environmental stress, such as treating the subset of crop plots with a treatment product, determining the crop injury for each of the exposed crop plots, determining the sensitivity on the basis of the crop injury for at least some, in particular all, of the exposed crop plots, wherein determining the crop injury and / or determining the sensitivity is based taking into account aerial image data of the unexposed crop plots, for example by using the unexposed crop plots as a reference.

18. The method of any of the preceding claims, comprising at least one of: selecting, among a plurality of candidate crops, a crop to be planted on the basis of the determined sensitivity, and / or selecting, among a plurality of candidate crops, a crop for breeding on the basis of the determined sensitivity.

19. A system (1) comprising a computing system (2) configured to carry out the method of any of the preceding claims.

20. The system of claim 19, comprising a selection unit (2a) configured to select a crop to be planted and / or a crop to be used for breeding on the basis of the determined sensitivity.

21. A computer program product comprising instructions which, when the program is executed by a computing system, cause the computing system to carry out the method of any of claims 1 to 18.

22. A computer-readable medium having stored thereon instructions which, when the program is executed by a computing system, cause the computing system to carry out the method of any of claims 1 to 18.

23. Use of the determined sensitivity of a crop to environmental stresses determined with the method of any one of claims 1 to 18 for trait development of crops, breeding, crop protection, and / or crop selection.

Images