Crop phytotoxicity detection method and device and storage medium

Abstract

The embodiment of the invention provides a crop phytotoxicity detection method and device and a storage medium. The method comprises the following steps: acquiring a multispectral image of crops in a planting area; carrying out splicing processing on the multispectral images to respectively obtain an RGB image, an NDVI image and a GNDVI image of the crop; extracting texture features of crops from the RGB image; performing deconvolution processing on the texture features to obtain plant pattern spots and a current growth period of the crops; determining a growth feature matrix of the crops according to the NDVI image and the GNDVI image; the phytotoxicity degree of the crops is determined according to the plant pattern spots, the current growth period and the growth vigor characteristic matrix, so that phytotoxicity detection of the crops is more accurate.

Description

Technical Field

[0001] This application relates to the field of crop pesticide damage detection technology, specifically to a crop pesticide damage detection method, device and storage medium. Background Technology

[0002] Currently, it is unavoidable to use pesticides during the growth of crops, such as herbicides. However, if too many pesticides are sprayed on crops in the planting area, it will lead to abnormal growth of crops, or even cause crops to fail to grow.

[0003] Current technology primarily relies on manual observation to detect pesticide damage in crops. However, in actual production, the accuracy of manual observation is very low, and it is easy to miss the optimal diagnosis time, resulting in pesticide damage not being treated in time and affecting crop growth. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, and storage medium for detecting crop pesticide damage, in order to solve the problem of inaccurate pesticide damage detection in the prior art.

[0005] To achieve the above objectives, the first aspect of this application provides a method for detecting crop pesticide damage, comprising:

[0006] Acquire multispectral images of crops within the planting area;

[0007] Multispectral images are stitched together to obtain RGB, NDVI, and GNDVI images of the crops.

[0008] Extracting texture features of crops from RGB images;

[0009] The texture features are deconvolutionally processed to obtain the crop plant patterns and current growth stage;

[0010] The crop growth characteristic matrix was determined based on NDVI and GNDVI images.

[0011] The degree of pesticide damage to crops is determined based on plant patterns, current growth stage, and growth characteristic matrix.

[0012] In this embodiment of the application, determining the crop growth feature matrix based on NDVI and GNDVI images includes: analyzing the NDVI and GNDVI images respectively to obtain multiple first growth assessment values ​​and multiple second growth assessment values; determining a first weight corresponding to each first growth assessment value and a second weight corresponding to each second growth assessment value based on the crop planting information and current accumulated temperature; and determining the crop growth feature matrix based on the multiple first growth assessment values, the multiple second growth assessment values, the first weight, and the second weight.

[0013] In this embodiment of the application, the analysis of NDVI images and GNDVI images to obtain multiple first growth assessment values ​​and multiple second growth assessment values ​​includes: for each first pixel in the NDVI image, determining the first growth assessment value corresponding to the first pixel based on the NDVI value of the first pixel, planting information, and current accumulated temperature; for each second pixel in the GNDVI image, determining the second growth assessment value corresponding to the second pixel based on the GNDVI value of the second pixel, planting information, and current accumulated temperature.

[0014] In this embodiment of the application, determining the crop growth feature matrix based on multiple first growth assessment values, multiple second growth assessment values, a first weight, and a second weight includes: determining a first product between the first growth assessment value corresponding to a first pixel at any position and the first weight; determining a second product between the second growth assessment value corresponding to a second pixel at any position and the second weight; determining the sum of the first product and the second product as the matrix component corresponding to any position; and determining the growth feature matrix based on the matrix components corresponding to all positions.

[0015] In this embodiment of the application, determining the degree of crop phytotoxicity based on plant patches, the current growth stage, and the growth characteristic matrix includes: determining the total number of effective pixels in the plant patches and the matrix formed by all pixel values ​​covering the crop in the plant patches; determining the coefficient corresponding to the current growth stage; and determining the degree of crop phytotoxicity based on the matrix formed by the total number of effective pixels, all pixel values ​​covering the crop in the plant patches, the coefficient, and the growth characteristic matrix.

[0016] In this embodiment of the application, the degree of phytotoxicity to the crop is determined by the following formula:

[0017]

[0018] Where Degree refers to the degree of pesticide damage to the crop, Period refers to the coefficient corresponding to the current growth stage of the crop, and M growth I(x,y) refers to the crop growth feature matrix, (x,y)∈A, where A is the pixel range covering the crop in the crop patch, I(x,y) is the pixel value of the pixel located in row x and column y in the crop patch, and pixels is the total number of valid pixels in the crop patch.

[0019] In this embodiment of the application, extracting the texture features of crops from an RGB image includes: inputting the RGB image into an image segmentation model to obtain the texture features of crops output by the image segmentation model.

[0020] In this embodiment of the application, the crop includes rice.

[0021] A second aspect of this application provides a crop pesticide damage detection device, comprising:

[0022] The memory is configured to store instructions;

[0023] The processor is configured to retrieve instructions from memory and, when executing the instructions, to implement the aforementioned crop pesticide damage detection method.

[0024] A third aspect of this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the aforementioned crop pesticide damage detection method.

[0025] The above technical solution determines the crop plant patch and current growth stage based on the texture features extracted from the RGB image, determines the crop growth feature matrix based on the NDVI and GNDVI images, and evaluates the degree of crop pesticide damage by combining the plant patch and the crop growth feature matrix at the current growth stage, making crop pesticide damage detection more accurate.

[0026] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0027] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings:

[0028] Figure 1 The schematic diagram illustrates a process flow diagram of a crop pesticide damage detection method according to an embodiment of this application;

[0029] Figure 2 A schematic flowchart of a crop pesticide damage detection method according to another embodiment of this application is shown.

[0030] Figure 3 The diagram illustrates the internal structure of a computer device according to an embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0032] It should be noted that if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0033] Figure 1 A schematic flowchart of a crop pesticide damage detection method according to an embodiment of this application is shown. Figure 1 As shown in one embodiment of this application, a method for detecting crop phytotoxicity is provided, comprising the following steps:

[0034] Step 101: Obtain multispectral images of crops within the planting area.

[0035] Crops are grown within the planting area. In this embodiment, the crop includes rice. The rice varieties may include various types, such as conventional indica rice, hybrid indica rice, conventional japonica rice, and hybrid japonica rice. Multispectral images of the crops can be acquired using a drone equipped with a multispectral image acquisition device. When acquiring multispectral images of the crops, the drone's flight altitude can be set within a suitable range, for example, a low altitude of 12m, to ensure clearer and more suitable multispectral images. The overlap of the drone's flight path can be set based on actual conditions, for example, it can be set to 75%. After acquiring the multispectral images, the multispectral image acquisition device can send the multispectral images to a processor. The processor can then acquire the multispectral images of the crops within the planting area.

[0036] Step 102: Perform stitching on the multispectral images to obtain the RGB image, NDVI image, and GNDVI image of the crop.

[0037] The processor can stitch together multispectral images to obtain RGB, NDVI, and GNDVI images of the crops. NDVI images represent the normalized difference in vegetation index (NDVI) of crops within the planting area, while GNDVI images represent the normalized green difference in vegetation index (GNDVI) of crops within the planting area. The normalized vegetation index is a crucial parameter reflecting crop growth and nutritional information. The GNDVI is more sensitive to green vegetation, allowing for a more accurate assessment of vegetation cover and growth status.

[0038] Step 103: Extract the texture features of the crop from the RGB image.

[0039] The processor can extract texture features of crops from RGB images. These texture features can include crop edge contour features, multi-scale features, local features, and structural features. The texture features of crops can be extracted using neural networks or image feature extraction algorithms.

[0040] In this embodiment of the application, extracting the texture features of crops from an RGB image includes: inputting the RGB image into an image segmentation model to obtain the texture features of crops output by the image segmentation model.

[0041] The processor can input RGB images into an image segmentation model to obtain the crop's texture features as output. The image segmentation model can be pre-trained; for example, multispectral images of the crop over a historical time period can be acquired, and the RGB images for that historical time period can be determined based on these multispectral images. A neural network can then be trained based on these RGB images to obtain the image segmentation model. This model can be used to extract the crop's texture features; image segmentation may include UNet.

[0042] Step 104: Perform deconvolution processing on the texture features to obtain the crop plant patches and current growth stage.

[0043] The processor can perform deconvolution processing on texture features to obtain crop plant patches and their current growth stage. Specifically, the processor can perform deconvolution processing on texture features using a crop detection model. The plant patch includes multiple pixels covering the crop, each pixel corresponding to a pixel value. The crop's growth stage can include sowing, emergence, tillering, jointing, booting, heading, grain-filling, and maturity stages.

[0044] Step 105: Determine the crop growth characteristic matrix based on the NDVI and GNDVI images.

[0045] The processor can determine the crop growth feature matrix based on NDVI and GNDVI images. In this embodiment, determining the crop growth feature matrix based on NDVI and GNDVI images includes: analyzing the NDVI and GNDVI images respectively to obtain multiple first growth assessment values ​​and multiple second growth assessment values; determining a first weight corresponding to each first growth assessment value and a second weight corresponding to each second growth assessment value based on the crop's planting information and current accumulated temperature; and determining the crop growth feature matrix based on the multiple first growth assessment values, the multiple second growth assessment values, the first weight, and the second weight.

[0046] The processor can analyze NDVI and GNDVI images separately to obtain multiple first growth assessment values ​​and multiple second growth assessment values. In this embodiment, analyzing NDVI and GNDVI images separately to obtain multiple first growth assessment values ​​and multiple second growth assessment values ​​includes: for each first pixel in the NDVI image, determining a first growth assessment value corresponding to the first pixel based on the NDVI value of the first pixel, planting information, and current accumulated temperature; and for each second pixel in the GNDVI image, determining a second growth assessment value corresponding to the second pixel based on the GNDVI value of the second pixel, planting information, and current accumulated temperature.

[0047] An NDVI image includes at least one first pixel. For each first pixel in the NDVI image, the processor can determine a first growth assessment value corresponding to the first pixel based on the NDVI value of the first pixel, crop planting information, and the current accumulated temperature of the crop.

[0048] The GNDVI image includes at least one second pixel. For each second pixel in the GNDVI image, the processor can determine a second growth assessment value corresponding to the second pixel based on the GNDVI value of the second pixel, planting information, and current accumulated temperature.

[0049] The processor can determine a first weight corresponding to each first growth assessment value and a second weight corresponding to each second growth assessment value based on crop planting information and current accumulated temperature. The processor can determine the crop growth feature matrix based on multiple first growth assessment values, multiple second growth assessment values, the first weight, and the second weight.

[0050] In this embodiment of the application, determining the crop growth feature matrix based on multiple first growth assessment values, multiple second growth assessment values, a first weight, and a second weight includes: determining a first product between the first growth assessment value corresponding to a first pixel at any position and the first weight; determining a second product between the second growth assessment value corresponding to a second pixel at any position and the second weight; determining the sum of the first product and the second product as the matrix component corresponding to any position; and determining the growth feature matrix based on the matrix components corresponding to all positions.

[0051] The processor can determine a first product between a first growth potential assessment value and a first weight corresponding to a first pixel at any location. The processor can determine a second product between a second growth potential assessment value and a second weight corresponding to a second pixel at any location. The processor can determine the sum of the first and second products as the matrix component corresponding to any location. The processor can determine the growth potential feature matrix based on the matrix components corresponding to all locations.

[0052] In one embodiment, the growth characteristic matrix can be determined based on the following formula:

[0053] M growth =αF(M ndvi (x,y))+βG(M gndvi (x,y))

[0054] Among them, M growth This refers to the growth feature matrix, where α is the first weight corresponding to each first growth assessment value in the NDVI image, F(M ndvi (x,y) refers to the first growth assessment value corresponding to each first pixel in the NDVI image, β refers to the second weight corresponding to each second growth assessment value in the GNDVI image, and G(M gndvi (x,y) refers to the second growth assessment value corresponding to each second pixel in the GNDVI image, where (x,y) refers to the coordinates of each pixel.

[0055] The above formula is constructed based on the normalized vegetation index and green light normalized vegetation index of crops of different varieties and planting areas, as well as the growth status of crops. Both α and β depend on the crop variety, transplanting time and accumulated temperature.

[0056] Step 106: Determine the degree of pesticide damage to the crop based on the plant patch, current growth stage, and growth characteristic matrix.

[0057] The processor can determine the degree of pesticide damage to crops based on plant patches, the current growth stage, and a growth characteristic matrix. In this embodiment, determining the degree of pesticide damage to crops based on plant patches, the current growth stage, and a growth characteristic matrix includes: determining the total number of effective pixels in the plant patches and the matrix formed by all pixel values ​​covering the crop in the plant patches; determining the coefficients corresponding to the current growth stage; and determining the degree of pesticide damage to crops based on the matrix formed by the total number of effective pixels, all pixel values ​​covering the crop in the plant patches, the coefficients, and the growth characteristic matrix.

[0058] The processor determines the total number of valid pixels in a plant patch. A plant patch comprises multiple pixels, each with a pixel value that can be set to 0 or 1. The total number of valid pixels refers to the number of valid pixels, which are pixels with a value of 1. If a pixel has a value of 1, it means that the pixel is within the crop plant area. If a pixel has a value of 0, it means that the pixel is outside the crop plant area, for example, in an area containing bare soil, water, or land.

[0059] The processor can determine the matrix formed by all the pixel values ​​covering the crop in a plant patch. The processor can also determine the coefficients corresponding to the current growth stage. That is, the coefficients differ depending on the crop's growth stage.

[0060] The processor can determine the degree of herbicide damage to crops based on the total number of effective pixels, a matrix composed of all pixel values ​​covering the crop in the plant patch, coefficients, and a growth characteristic matrix. Specifically, the processor can determine the sum of the growth levels of the crop pixels within the planting area based on the matrix composed of all pixel values ​​covering the crop in the plant patch, coefficients, and a growth characteristic matrix. Then, it can combine this with the total number of effective pixels to determine the degree of herbicide damage. The degree of herbicide damage can be set to different levels, such as high herbicide damage, moderate herbicide damage, and low herbicide damage.

[0061] In this embodiment of the application, the degree of phytotoxicity to crops is determined by the following formula:

[0062]

[0063] Where Degree refers to the degree of pesticide damage to the crop, Period refers to the coefficient corresponding to the current growth stage of the crop, and M growth I(x,y) refers to the crop growth feature matrix, (x,y)∈A, where A is the pixel range covering the crop in the crop patch, I(x,y) is the pixel value of the pixel located in row x and column y in the crop patch, and pixels is the total number of valid pixels in the crop patch.

[0064] like Figure 2 As shown, using rice as an example, a flowchart of another method for detecting crop pesticide damage is provided.

[0065] When detecting pesticide damage in rice, drones can be used to collect multispectral images of the planting area at a low altitude of 12m. The overlap of the drone's flight path can be set to 75%. The multispectral images can then be stitched together to obtain RGB, NDVI, and GNDVI images.

[0066] Using the constructed Unet network structure, texture features of rice are extracted from RGB images. Functional relationships between NDVI, GNDVI, and growth vigor are established for different rice varieties and planting areas. A rice plant detection model is constructed, and deconvolution calculations are performed on the texture features to obtain rice plant patches and growth stages. The rice growth feature matrix is ​​determined using NDVI and GNDVI images, combined with the constructed functional relationships.

[0067] A rice herbicide damage assessment model was constructed, and the degree of herbicide damage was determined by combining the rice growth characteristic matrix, growth stage, and plant patch data. Specifically, the number of effective pixels in the plant patch data can be counted, and the degree of herbicide damage can be determined based on the number of effective pixels, the coefficient corresponding to the growth stage, the rice growth characteristic matrix, and the matrix corresponding to the plant patch data.

[0068] The above technical solution determines the crop plant patch and current growth stage based on the texture features extracted from the RGB image, determines the crop growth feature matrix based on the NDVI and GNDVI images, and evaluates the degree of crop pesticide damage by combining the plant patch and the crop growth feature matrix at the current growth stage, making crop pesticide damage detection more accurate.

[0069] Figure 1 and 2 This is a flowchart illustrating a crop phytotoxicity detection method in one embodiment. It should be understood that, although... Figure 1 and 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 and 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0070] In one embodiment, a crop pesticide damage detection device is provided, comprising:

[0071] The memory is configured to store instructions;

[0072] The processor is configured to retrieve instructions from memory and, when executing the instructions, to implement the aforementioned crop pesticide damage detection method.

[0073] In one embodiment, a storage medium is provided on which a program is stored, which, when executed by a processor, implements the above-described crop pesticide damage detection method.

[0074] In one embodiment, a processor is provided for running a program, wherein the program executes the above-described crop pesticide damage detection method during runtime.

[0075] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor A01, a network interface A02, a memory (not shown), and a database (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02, and a database (not shown). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A04. The database stores data such as the degree of crop pesticide damage. The network interface A02 is used for communication with external terminals via a network connection. When the processor A01 executes the computer program B02, it implements a crop pesticide damage detection method.

[0076] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0077] This application provides an apparatus including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring multispectral images of crops within a planting area; stitching the multispectral images to obtain RGB, NDVI, and GNDVI images of the crops; extracting texture features of the crops from the RGB images; performing deconvolution processing on the texture features to obtain plant patches and the current growth stage of the crops; determining the crop growth feature matrix based on the NDVI and GNDVI images; and determining the degree of pesticide damage to the crops based on the plant patches, the current growth stage, and the growth feature matrix.

[0078] In one embodiment, determining the crop growth feature matrix based on NDVI and GNDVI images includes: analyzing the NDVI and GNDVI images respectively to obtain multiple first growth assessment values ​​and multiple second growth assessment values; determining a first weight corresponding to each first growth assessment value and a second weight corresponding to each second growth assessment value based on crop planting information and current accumulated temperature; and determining the crop growth feature matrix based on the multiple first growth assessment values, the multiple second growth assessment values, the first weight, and the second weight.

[0079] In one embodiment, analyzing the NDVI image and the GNDVI image respectively to obtain multiple first growth assessment values ​​and multiple second growth assessment values ​​includes: for each first pixel in the NDVI image, determining the first growth assessment value corresponding to the first pixel based on the NDVI value of the first pixel, planting information, and current accumulated temperature; and for each second pixel in the GNDVI image, determining the second growth assessment value corresponding to the second pixel based on the GNDVI value of the second pixel, planting information, and current accumulated temperature.

[0080] In one embodiment, determining the crop growth feature matrix based on a plurality of first growth assessment values, a plurality of second growth assessment values, a first weight, and a second weight includes: determining a first product between the first growth assessment value corresponding to a first pixel at any location and the first weight; determining a second product between the second growth assessment value corresponding to a second pixel at any location and the second weight; determining the sum of the first product and the second product as the matrix component corresponding to any location; and determining the growth feature matrix based on the matrix components corresponding to all locations.

[0081] In one embodiment, determining the degree of crop phytotoxicity based on plant patches, the current growth stage, and a growth characteristic matrix includes: determining the total number of effective pixels in the plant patches and the matrix formed by all pixel values ​​covering the crop in the plant patches; determining the coefficient corresponding to the current growth stage; and determining the degree of crop phytotoxicity based on the matrix formed by the total number of effective pixels, all pixel values ​​covering the crop in the plant patches, the coefficient, and the growth characteristic matrix.

[0082] In one embodiment, the degree of phytotoxicity to crops is determined by the following formula:

[0083]

[0084] Where Degree refers to the degree of pesticide damage to the crop, Period refers to the coefficient corresponding to the current growth stage of the crop, and M growth I(x,y) refers to the crop growth feature matrix, (x,y)∈A, where A is the pixel range covering the crop in the crop patch, I(x,y) is the pixel value of the pixel located in row x and column y in the crop patch, and pixels is the total number of valid pixels in the crop patch.

[0085] In one embodiment, extracting texture features of crops from an RGB image includes: inputting the RGB image into an image segmentation model to obtain texture features of crops output by the image segmentation model.

[0086] In one embodiment, the crop includes rice.

[0087] This application also provides a computer program product that, when executed on a data processing device, is adapted to execute a program that initializes the above-described crop pesticide damage detection method steps.

[0088] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0093] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0094] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0095] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0096] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for detecting crop pesticide damage, characterized in that, include: Acquire multispectral images of crops within the planting area; The multispectral images are stitched together to obtain RGB, NDVI, and GNDVI images of the crop, respectively. Extract the texture features of the crop from the RGB image; The texture features are deconvolutionally processed to obtain the plant patch and current growth stage of the crop. The growth characteristic matrix of the crop is determined based on the NDVI image and the GNDVI image; The degree of pesticide damage to the crop is determined based on the plant patch, the current growth stage, and the growth characteristic matrix.

2. The crop pesticide damage detection method according to claim 1, characterized in that, The step of determining the crop growth feature matrix based on the NDVI image and the GNDVI image includes: The NDVI image and the GNDVI image are analyzed respectively to obtain multiple first growth assessment values ​​and multiple second growth assessment values; Based on the crop planting information and the current accumulated temperature, a first weight corresponding to each first growth assessment value and a second weight corresponding to each second growth assessment value are determined respectively. The crop growth characteristic matrix is ​​determined based on the plurality of first growth assessment values, the plurality of second growth assessment values, the first weight, and the second weight.

3. The crop pesticide damage detection method according to claim 2, characterized in that, The analysis of the NDVI image and the GNDVI image to obtain multiple first growth assessment values ​​and multiple second growth assessment values ​​includes: For each first pixel in the NDVI image, a first growth assessment value corresponding to the first pixel is determined based on the NDVI value of the first pixel, the planting information, and the current accumulated temperature. For each second pixel in the GNDVI image, a second growth assessment value corresponding to the second pixel is determined based on the GNDVI value of the second pixel, the planting information, and the current accumulated temperature.

4. The crop pesticide damage detection method according to claim 2, characterized in that, The step of determining the crop growth characteristic matrix based on the plurality of first growth assessment values, the plurality of second growth assessment values, the first weight, and the second weight includes: Determine the first product between the first growth assessment value corresponding to the first pixel point located at any position and the first weight; Determine the second product between the second growth potential evaluation value corresponding to the second pixel point located at any of the above positions and the second weight; The sum of the first product and the second product is determined as the matrix component corresponding to any of the positions; The growth characteristic matrix is ​​determined based on the matrix components corresponding to all positions.

5. The crop pesticide damage detection method according to claim 1, characterized in that, The determination of the degree of pesticide damage to the crop based on the plant patch, the current growth stage, and the growth characteristic matrix includes: Determine the total number of valid pixels in the plant patch, and the matrix formed by all pixel values ​​in the plant patch covering the crop; Determine the coefficient corresponding to the current reproductive period; The degree of pesticide damage to the crop is determined based on the total number of effective pixels, the matrix formed by all pixel values ​​covering the crop in the plant patch, the coefficients, and the growth characteristic matrix.

6. The crop pesticide damage detection method according to claim 5, characterized in that, The degree of phytotoxicity to the crop is determined by the following formula: Where Degree refers to the degree of pesticide damage to the crop, Period refers to the coefficient corresponding to the current growth stage of the crop, and M growth I(x,y) refers to the crop growth feature matrix, (x,y)∈A, where A is the pixel range covering the crop in the crop patch, I(x,y) is the pixel value of the pixel located in row x and column y in the crop patch, and pixels is the total number of valid pixels in the crop patch.

7. The crop pesticide damage detection method according to claim 1, characterized in that, The extraction of texture features of the crop from the RGB image includes: The RGB image is input into the image segmentation model to obtain the texture features of the crop output by the image segmentation model.

8. The crop pesticide damage detection method according to any one of claims 1 to 7, characterized in that, The crop mentioned includes rice.

9. A crop pesticide damage detection device, characterized in that, The device includes: The memory is configured to store instructions; The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the crop pesticide damage detection method according to any one of claims 1 to 8.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform the crop pesticide damage detection method according to any one of claims 1 to 8.

Images