Cotton verticillium wilt resistance detection method based on unmanned aerial vehicle chlorophyll detection model

Abstract

The invention provides a cotton verticillium wilt resistance detection method based on an unmanned aerial vehicle chlorophyll detection model, relates to the technical field of data detection, and solves the technical problem of low cotton verticillium wilt resistance detection efficiency. The method comprises the following steps: acquiring RGB data and multispectral data of a specified field area of cotton through image acquisition equipment arranged on an unmanned aerial vehicle; based on the RGB digital orthoimage, the multispectral digital orthoimage and chlorophyll content data measured on site by different varieties of cotton, utilizing a machine learning algorithm to construct chlorophyll detection models of chlorophyll of different varieties of cotton in different growth periods by adjusting initial model parameters; based on the optimal chlorophyll detection model and the verticillium wilt disease index data, K-means clustering analysis and GIS space analysis are carried out, chlorophyll change rule data and verticillium wilt resistance levels of different varieties of cotton are obtained, and the verticillium wilt resistance levels of different varieties of cotton serve as cotton verticillium wilt resistance detection results.

Description

Technical Field

[0001] The present application relates to the field of data detection technology, and in particular to a method for detecting cotton Verticillium wilt resistance based on an unmanned aerial vehicle (UAV) chlorophyll detection model. Background Art

[0002] The production of many crops around the world, including cotton, potatoes, tomatoes, peppers, etc., is currently under serious threat from Verticillium wilt, causing significant economic losses. Cotton is recognized as an important economic crop worldwide and is also the main source of natural fibers for the textile industry. Verticillium wilt is the "cancer" of cotton. It is widely distributed and highly pathogenic. It is one of the most destructive diseases and can significantly reduce cotton yield and fiber quality. Currently, the commonly used prevention and control methods are chemical control and biological control. However, due to the difficulty in accurately determining the area and severity of Verticillium wilt, chemical control by large-scale spraying of pesticides often causes waste of manpower and material resources and environmental pollution. Biological control has the characteristics of high safety characteristics, but its disadvantages such as high cost, long time consumption, and slow effect limit its widespread application in field agriculture.

[0003] At present, manual observation, as a traditional method for detecting field crop diseases, mainly relies on visual inspection and personal experience, so it has the disadvantage of low detection efficiency. The indoor dynamic observation system can effectively and automatically investigate the severity of cotton Verticillium wilt, but it cannot be applied to Verticillium wilt detection at the field scale, resulting in low efficiency of existing cotton Verticillium wilt resistance detection. Summary of the invention

[0004] The purpose of the present invention is to provide a cotton Verticillium wilt resistance detection method based on an unmanned aerial vehicle chlorophyll detection model to solve the technical problem of low efficiency in cotton Verticillium wilt resistance detection.

[0005] In a first aspect, the present application provides a method for detecting cotton Verticillium wilt resistance based on an unmanned aerial vehicle chlorophyll detection model, the method comprising:

[0006] The image acquisition equipment installed on the drone is used to obtain RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth stages of cotton in the field;

[0007] Using a chlorophyll analyzer, field-measured chlorophyll content data of different varieties of cotton are collected in the designated infected field area and the designated uninfected field area; the time difference between the collection time corresponding to the drone and the collection time corresponding to the chlorophyll analyzer is less than a designated time period;

[0008] Calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt severity disease grading standard; wherein the disease index is used to characterize the disease severity of Verticillium wilt;

[0009] The RGB digital orthophoto in the RGB data is input into an image automatic recognition and classification model through an image classification code, and the image classification is automatically performed based on the RGB digital orthophoto and the cotton canopy range is divided through the image automatic recognition and classification model, and the cotton canopy vector boundaries of multiple varieties of cotton in different growth stages are extracted based on the image classification results and the cotton canopy range division results;

[0010] Calculating multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extracting the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and performing a correlation analysis based on the mean of the vegetation indices and the disease index of cotton Verticillium wilt, to obtain the optimal vegetation index sensitive to cotton Verticillium wilt;

[0011] Based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average of the optimal vegetation index, and the chlorophyll content data of different varieties of cotton measured on the spot, a chlorophyll detection model of different varieties of cotton chlorophyll at different growth stages is constructed and optimized by adjusting initial model parameters using a machine learning algorithm to obtain an optimal chlorophyll detection model;

[0012] Based on the optimal chlorophyll detection model, K-means cluster analysis and GIS spatial analysis are performed to obtain chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of the different varieties of cotton;

[0013] According to the correlation relationship, the degree of correlation between chlorophyll and Verticillium wilt in different growth periods is determined, and based on the chlorophyll content data of the different varieties of cotton measured in the field, the resistance levels of the different varieties of cotton to Verticillium wilt are determined through the degree of correlation between chlorophyll and Verticillium wilt, and the resistance levels of the different varieties of cotton to Verticillium wilt are used as the results of cotton Verticillium wilt resistance detection.

[0014] In a possible implementation, the step of calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt disease severity classification standard includes:

[0015] According to the designated cotton Verticillium wilt severity classification standard, the disease index of cotton Verticillium wilt of different varieties during the cotton Verticillium wilt outbreak period is calculated by the following formula:

[0016]

[0017] Among them, Zmi is the number of different cotton Verticillium wilt severity levels, n is the number of the highest cotton Verticillium wilt severity level, Z pi is the number of plants in each disease severity level of cotton Verticillium wilt, and i is the disease severity level of cotton Verticillium wilt.

[0018] In a possible implementation, after acquiring the RGB data and multispectral data of the designated infected field area and the designated uninfected field area at different key growth periods of the cotton field by the image acquisition device arranged on the drone, the method further includes:

[0019] Image preprocessing is performed on the RGB digital orthophoto in the RGB data and the multispectral digital orthophoto in the multispectral data, wherein the image preprocessing includes at least one of the following: radiation calibration, geometric correction, image alignment, camera alignment optimization, dense point cloud generation, depth map creation, generation of the RGB digital orthophoto model, and generation of the multispectral digital orthophoto model.

[0020] In a possible implementation, the plurality of vegetation indices include at least one of the following: normalized difference vegetation index NDVI, difference vegetation index DVI, green-red vegetation index GRVI, infrared percentage vegetation index IPVI, and green normalized difference vegetation index GNDVI; the method further includes:

[0021] Acquire a multispectral digital orthophoto after image preprocessing in the multispectral data; wherein the multispectral digital orthophoto after image preprocessing is an image synthesized by green light, red light, red edge and near infrared bands;

[0022] The image synthesized by the green light, red light, red edge and near infrared bands is band-splitting based on the multiple vegetation indices sensitive to chlorophyll to obtain multiple single bands after band splitting, and the multiple single bands are used for calculation to obtain the optimal vegetation index sensitive to cotton Verticillium wilt.

[0023] In a possible implementation, extracting the mean value of the optimal vegetation index corresponding to different varieties of cotton according to the cotton canopy vector boundary includes:

[0024] The mean values ​​of the optimal vegetation index corresponding to different varieties of cotton are extracted using the QGIS program according to the cotton canopy vector boundary and the optimal vegetation index.

[0025] In a possible implementation, the chlorophyll detection model of different varieties of cotton chlorophyll in different growth periods is constructed and optimized by adjusting the initial model parameters based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the mean of the optimal vegetation index, and the chlorophyll content data of different varieties of cotton measured in the field, using a machine learning algorithm to obtain the optimal chlorophyll detection model, including:

[0026] Based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index and the mean of the optimal vegetation index, three machine learning algorithms are used to construct and optimize the chlorophyll detection model of different varieties of cotton chlorophyll in different growth periods by adjusting the initial model parameters, and the accuracy of the chlorophyll detection model is detected by calculating the determination coefficient, root mean square error and mean absolute error to obtain the optimal chlorophyll detection model, wherein the higher the value of the determination coefficient and the lower the values ​​of the root mean square error and the mean absolute error, the higher the accuracy of the chlorophyll detection result corresponding to the chlorophyll detection model; the three machine learning algorithms include at least random forest, support vector machine and BP neural network.

[0027] In a possible implementation, the K-means cluster analysis and GIS spatial analysis are performed based on the optimal chlorophyll detection model to obtain chlorophyll variation law data and chlorophyll spatial distribution law data of different varieties of cotton, including:

[0028] K-means cluster analysis is performed based on the optimal chlorophyll detection model to obtain multiple clusters with different characteristics. Based on the multiple clusters, GIS spatial analysis is performed to obtain the chlorophyll change law and spatial distribution law of different varieties of cotton.

[0029] In a possible implementation, determining the correlation between the chlorophyll variation law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of different varieties of cotton includes:

[0030] The chlorophyll change law data and the chlorophyll spatial distribution law data are compared with the spatial distribution laws of the incidence of Verticillium wilt in the different varieties of cotton, and the correlation between chlorophyll and the disease index of Verticillium wilt in the different varieties of cotton is analyzed based on the comparison results, so as to detect the disease resistance characteristics of different varieties of cotton by measuring the chlorophyll content in the cotton canopy.

[0031] In the second aspect, the present application provides a cotton Verticillium wilt resistance detection device based on a drone chlorophyll detection model, comprising:

[0032] An acquisition module is used to acquire RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth periods of cotton in the field through an image acquisition device set on the UAV;

[0033] A collection module is used to collect chlorophyll content data of different varieties of cotton measured in the designated infected field area and the designated uninfected field area respectively by using a chlorophyll analyzer; the time difference between the collection time corresponding to the drone and the collection time corresponding to the chlorophyll analyzer is less than a specified time period;

[0034] A calculation module, for calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt severity disease classification standard; wherein the disease index is used to characterize the disease severity of Verticillium wilt;

[0035] An extraction module is used to input the RGB digital orthophoto in the RGB data into an image automatic recognition and classification model through an image classification code, automatically perform image classification based on the RGB digital orthophoto and divide the cotton canopy range through the image automatic recognition and classification model, and extract cotton canopy vector boundaries of multiple varieties of cotton in different growth stages based on the image classification results and the cotton canopy range division results;

[0036] An analysis module, for calculating multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extracting the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and performing a correlation analysis based on the mean of the vegetation indices and the disease index of the cotton Verticillium wilt, to obtain the optimal vegetation index sensitive to cotton Verticillium wilt;

[0037] A construction module is used to construct and optimize the chlorophyll detection model of different varieties of cotton chlorophyll in different growth stages by adjusting the initial model parameters based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average of the optimal vegetation index and the chlorophyll content data of different varieties of cotton measured on the spot, so as to obtain the optimal chlorophyll detection model;

[0038] The first determination module is used to perform K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of the different varieties of cotton;

[0039] The second determination module is used to determine the degree of association between chlorophyll in different growth periods and Verticillium wilt according to the association relationship, and to determine the resistance level of different varieties of cotton to Verticillium wilt based on the chlorophyll content data of the different varieties of cotton measured in the field through the degree of association between chlorophyll and Verticillium wilt, and to use the resistance level of the different varieties of cotton to Verticillium wilt as the cotton Verticillium wilt resistance detection result.

[0040] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the method described in the first aspect.

[0042] This application brings the following beneficial effects:

[0043] The present application provides a method for detecting resistance to Verticillium wilt in cotton based on a chlorophyll detection model provided by an unmanned aerial vehicle. The method can obtain RGB data and multispectral data of designated infected field areas and designated uninfected field areas in different key growth periods of cotton in the field through an image acquisition device arranged on the unmanned aerial vehicle. The chlorophyll content data of different varieties of cotton measured in the field are collected in the designated infected field areas and designated uninfected field areas respectively through a chlorophyll analyzer. The time difference between the collection time corresponding to the unmanned aerial vehicle and the collection time corresponding to the chlorophyll analyzer is less than a specified time period. The disease index of Verticillium wilt in different varieties of cotton during the onset period of Verticillium wilt in cotton is calculated according to the specified cotton Verticillium wilt severity disease grading standard, wherein the disease index is used to characterize The severity of Verticillium wilt is determined by inputting the RGB digital orthophoto in the RGB data into the image automatic recognition and classification model through the image classification code. The image classification is automatically performed based on the RGB digital orthophoto and the cotton canopy range is divided through the image automatic recognition and classification model. The cotton canopy vector boundaries of multiple varieties of cotton in different growth stages are extracted based on the image classification results and the cotton canopy range division results. Multiple chlorophyll-sensitive vegetation indices are calculated based on the multispectral digital orthophoto in the multispectral data. The mean of the vegetation index corresponding to different varieties of cotton is extracted according to the cotton canopy vector boundary. The correlation analysis is performed based on the mean of the vegetation index and the disease index of cotton Verticillium wilt, and the optimal vegetation index sensitive to cotton Verticillium wilt is obtained. RGB digital orthophotos, multispectral digital orthophotos, optimal vegetation index, the mean of the optimal vegetation index and the chlorophyll content data of different varieties of cotton measured on the spot, and using machine learning algorithms to adjust the initial model parameters to construct and optimize the chlorophyll detection model of different varieties of cotton chlorophyll in different growth periods, and obtain the optimal chlorophyll detection model, and perform K-means clustering analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain the chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of different varieties of cotton Verticillium wilt, and determine the disease index of different growth periods according to the correlation. The correlation between chlorophyll and Verticillium wilt in the period was determined, and based on the chlorophyll content data of different varieties of cotton measured in the field, the resistance level of different varieties of cotton to Verticillium wilt was determined through the correlation between chlorophyll and Verticillium wilt, and the resistance level of different varieties of cotton to Verticillium wilt was used as the detection result of cotton Verticillium wilt resistance. In this scheme, before using chlorophyll to detect the Verticillium wilt resistance of multiple varieties of cotton, drone data collection, chlorophyll measured data collection and Verticillium wilt disease index collection were first used. After analyzing and processing the collected data, a drone chlorophyll detection model was constructed and optimized using a machine learning algorithm. Based on this model, the purpose of detecting the Verticillium wilt resistance level of multiple varieties of cotton can be achieved by simply measuring the chlorophyll content data of the cotton canopy, and,With the help of high-throughput and high-precision UAV remote sensing platforms, the area and extent of Verticillium wilt in the field can be detected in a timely and accurate manner, and the resistance level of different cotton varieties to Verticillium wilt can be effectively detected, which improves the efficiency of cotton Verticillium wilt resistance detection and solves the technical problem of low efficiency of cotton Verticillium wilt resistance detection.

[0044] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 A schematic diagram of a process for detecting cotton Verticillium wilt resistance based on a drone chlorophyll detection model provided in an embodiment of the present application;

[0047] Figure 2 Another schematic diagram of the process of detecting cotton Verticillium wilt resistance based on the drone chlorophyll detection model provided in the embodiment of the present application;

[0048] Figure 3 A schematic diagram of the structure of a cotton Verticillium wilt resistance detection device based on a drone chlorophyll detection model provided in an embodiment of the present application;

[0049] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0051] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0052] At present, the existing cotton Verticillium wilt resistance detection efficiency is low. Based on this, the embodiment of the present application provides a cotton Verticillium wilt resistance detection method based on a drone chlorophyll detection model, which can solve the technical problem of low cotton Verticillium wilt resistance detection efficiency.

[0053] The embodiments of the present invention are further described below in conjunction with the accompanying drawings.

[0054] Figure 1 The present invention provides a flowchart of a method for detecting cotton Verticillium wilt resistance based on a drone chlorophyll detection model. Figure 1 As shown, the method includes:

[0055] Step S110, obtaining RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth stages of cotton in the field through an image acquisition device arranged on the drone.

[0056] In this step, if Figure 2 As shown in the figure, RGB and multispectral data of infected and uninfected areas of cotton at different key growth stages were obtained in the field through UAV remote sensing technology. It should be noted that by setting up the Verticillium wilt infected area and the uninfected control area, it is helpful to accurately analyze the dynamic changes of leaf chlorophyll in different varieties of cotton at multiple key growth periods, under Verticillium wilt infected and uninfected conditions.

[0057] As a possible implementation method, the RGB and multi-spectral data of infected and uninfected areas of cotton in different key growth stages (key growth stages include bud stage, flowering stage, full flowering stage, boll stage, full boll stage) are obtained in the field through the use of UAV remote sensing technology and real-time dynamic differential positioning module (RTK). The user needs to determine the field scanning range of the UAV. The present invention sets the flight route according to the size of the user's plot, ensuring a 70% lateral overlap rate and an 80% forward overlap rate.

[0058] More specifically, before acquiring multispectral data from drones, since the numerical characteristics of the drone sensor's response to light are related to the ambient light intensity and camera parameters, it is necessary to first perform radiation correction using a reflector with a reflectivity of 95% under the same ambient light intensity. The correction process is performed based on the Pix4D software platform during subsequent data preprocessing to obtain more accurate spectral information.

[0059] Step S120, using a chlorophyll analyzer to collect chlorophyll content data of different varieties of cotton measured in the designated infected field area and the designated uninfected field area.

[0060] It should be noted that the time difference between the collection time corresponding to the UAV and the collection time corresponding to the chlorophyll analyzer is less than the specified time period.

[0061] like Figure 2 As shown in the figure, while the drone obtains RGB and multispectral data of cotton at different growth stages in the field, it is necessary to measure the chlorophyll content data of the leaves, that is, using the SPAD-502PLUS chlorophyll analyzer, select different varieties of cotton in the infected area and the uninfected area to measure the chlorophyll content data on the spot.

[0062] In an optional embodiment, on the same day as the drone data collection, a SPAD-502PLUS chlorophyll analyzer is used to measure the chlorophyll content data of the canopy leaves of different varieties of cotton in the infected and uninfected areas, and use the measured data as the training set data, and another part as the measured data of the test set; the chlorophyll spectrometer needs to be spectrally corrected using a filter before use to prevent excessive errors from affecting the measurement accuracy of the chlorophyll content data.

[0063] Step S130, calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt outbreak according to the designated cotton Verticillium wilt severity disease grading standard.

[0064] The disease index is used to characterize the severity of Verticillium wilt. Figure 2 As shown, the disease index representing the severity of Verticillium wilt is analyzed in the later stage of Verticillium wilt, that is, the disease index of Verticillium wilt of different varieties of cotton in the region is calculated according to the cotton Verticillium wilt severity grading standard.

[0065] As an optional implementation, the disease index of Verticillium wilt of different varieties of cotton during the Verticillium wilt outbreak period is calculated according to the specified cotton Verticillium wilt severity disease grading standard by the following formula:

[0066]

[0067] Among them, Z miis the number of different cotton Verticillium wilt severity levels, n is the number of the highest cotton Verticillium wilt severity level, Z pi is the number of plants with each severity level of cotton Verticillium wilt disease, and i is the severity level of cotton Verticillium wilt disease. The above formula can more accurately calculate the disease index of Verticillium wilt disease of different varieties of cotton.

[0068] Step S140, input the RGB digital orthophoto in the RGB data into the image automatic recognition and classification model through the image classification code, automatically perform image classification and divide the cotton canopy range based on the RGB digital orthophoto through the image automatic recognition and classification model, and extract the cotton canopy vector boundaries of multiple varieties of cotton in different growth periods based on the image classification results and the cotton canopy range division results.

[0069] It should be noted that the digital orthophoto model is divided into RGB true color digital orthophoto and multispectral digital orthophoto according to different data sources. Figure 2 As shown in the figure, RGB true color digital orthophotos are used to extract cotton canopy boundaries, and multispectral digital orthophotos are used to extract vegetation indices and build chlorophyll detection models.

[0070] As an example, the code is used to automatically classify and divide RGB digital orthophotos and extract the canopy range of multiple varieties of cotton at different growth stages. Specifically, based on the Python image classification code, the RGB digital orthophoto is input, the canopy range is automatically divided and the canopy boundaries of multiple varieties of cotton at different growth stages are extracted, and saved in shp format for subsequent extraction of spectral information of different varieties.

[0071] In practical applications, digital orthoimage models are divided into RGB true color digital orthoimages and multispectral digital orthoimages according to different data sources. Based on the Python image classification code, the RGB digital orthoimage is input to automatically divide and extract the canopy range of multiple varieties of cotton in different growth periods, and save it in shp format for subsequent extraction of spectral information of each variety.

[0072] Step S150, calculating multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extracting the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and performing a correlation analysis between the mean of the vegetation index and the disease index of cotton Verticillium wilt, to obtain the optimal vegetation index sensitive to cotton Verticillium wilt.

[0073] In one possible implementation, multiple vegetation indices sensitive to chlorophyll are calculated based on multispectral digital orthophotos, and the mean values ​​of vegetation indices corresponding to different varieties of cotton are extracted using cotton canopy vector boundaries to perform correlation analysis and determine the optimal vegetation index sensitive to Verticillium wilt in cotton.

[0074] As an example, the plurality of vegetation indices include at least one of the following: normalized difference vegetation index NDVI, difference vegetation index DVI, green-red vegetation index GRVI, infrared percentage vegetation index IPVI, and green normalized difference vegetation index GNDVI; the method further includes the following steps:

[0075] A multispectral digital orthophoto is obtained after image preprocessing in the multispectral data; wherein the multispectral digital orthophoto after image preprocessing is an image synthesized by green light, red light, red edge and near infrared bands; based on multiple vegetation indices sensitive to chlorophyll, the image synthesized by green light, red light, red edge and near infrared bands is band-splitting to obtain multiple single bands after band splitting, and multiple single bands are used for calculation to obtain the optimal vegetation index sensitive to cotton Verticillium wilt.

[0076] In a possible implementation, band calculation can be performed on the ENVI software platform, and the vegetation index obtained by spectral calculation is saved in TIFF format for extracting vegetation indexes of different species at different periods.

[0077] For example, Figure 2 As shown, based on the multiple vegetation indices of chlorophyll sensitivity, the pre-processed multispectral digital orthophoto is used to calculate the correlation analysis to determine the optimal vegetation index (i.e., calculate the vegetation index) sensitive to cotton Verticillium wilt. And based on the vector boundaries of different varieties of cotton obtained in the above step S140, the mean values ​​of the vegetation index of different varieties of cotton are extracted.

[0078] For example, various vegetation indices based on chlorophyll sensitivity, such as NDVI, DVI, GRVI, IPVI, GNDVI, etc., are calculated using pre-processed multispectral digital orthophotos. Multispectral digital orthophotos are stacked images of green light, red light, red edge, and near-infrared bands. Before calculating the vegetation index, band splitting is required, and the vegetation index is calculated using multiple single bands.

[0079] As an optional implementation, the above process of extracting the mean of the optimal vegetation index corresponding to different varieties of cotton according to the cotton canopy vector boundary can specifically include the following steps: extracting the mean of the optimal vegetation index corresponding to different varieties of cotton according to the cotton canopy vector boundary and the optimal vegetation index using the QGIS program. For example, based on the vector boundary of different varieties of cotton obtained in step S140 above, the vector boundary of different varieties of cotton is extracted using the QGIS software platform, and tiff format images of different periods are input, and the vegetation index mean of different varieties of cotton is output.

[0080] Step S160, based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the mean of the optimal vegetation index and the chlorophyll content data of different varieties of cotton measured in the field, a chlorophyll detection model of different varieties of cotton at different growth stages is constructed and optimized by adjusting the initial model parameters using a machine learning algorithm to obtain an optimal chlorophyll detection model.

[0081] In an optional embodiment, if Figure 2 As shown in the figure, based on three machine learning algorithms, namely random forest, support vector machine and BP neural network, the detection model of chlorophyll of different varieties of cotton at different growth stages was constructed and optimized by adjusting the model parameters. 2 ), root mean square error (RMSE) and mean absolute error (MAE) detection model accuracy, and the optimal chlorophyll detection model is obtained, R 2 The higher the value, the lower the RMSE and MAE values, indicating that the model estimates the SPAD value with higher accuracy and precision, thus achieving the improvement of the accuracy of the chlorophyll prediction model for the entire growth period through machine learning algorithms.

[0082] As an example, based on RGB digital orthophotos, multispectral digital orthophotos, optimal vegetation index and the mean of the optimal vegetation index, three machine learning algorithms are used to construct and optimize the chlorophyll detection model of different varieties of cotton in different growth periods by adjusting the initial model parameters, and the accuracy of the chlorophyll detection model is detected by calculating the determination coefficient, root mean square error and mean absolute error to obtain the optimal chlorophyll detection model, wherein the higher the value of the determination coefficient and the lower the value of the root mean square error and the mean absolute error, the higher the accuracy of the chlorophyll detection result corresponding to the chlorophyll detection model; the three machine learning algorithms include at least random forest, support vector machine and BP neural network.

[0083] For example, based on different machine learning methods (such as random forest, support vector machine and BP neural network algorithm), by adjusting the model parameters, the detection model of chlorophyll of different varieties of cotton at different growth stages was constructed and optimized, and the determination coefficient (R 2), root mean square error (RMSE) and mean absolute error (MAE) detection model accuracy, and the optimal chlorophyll detection model is obtained, R 2 The higher the value, the lower the RMSE and MAE values, indicating that the model has higher accuracy and precision in estimating the SPAD value.

[0084] By using machine learning such as random forest, support vector machine and BP neural network algorithm, a multi-growth period chlorophyll prediction model was constructed and optimized, and a reliable vegetation index that can be used to detect chlorophyll levels under Verticillium wilt and normal conditions can be identified. This fills the gap in the study of the impact of Verticillium wilt on the chlorophyll content of different varieties of cotton in field agriculture.

[0085] Step S170, performing K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain chlorophyll variation law data and chlorophyll spatial distribution law data of different varieties of cotton, and determining the correlation between the chlorophyll variation law data, chlorophyll spatial distribution law data and the disease index of Verticillium wilt of different varieties of cotton.

[0086] It should be noted that the optimized chlorophyll detection model was used for K-means cluster analysis and GIS spatial analysis to explore the chlorophyll change and spatial distribution patterns of different varieties of cotton, and further analyze the association with the Verticillium wilt disease index.

[0087] For example, Figure 2 As shown in the figure, based on the chlorophyll detection model optimized by machine learning, K-means cluster analysis and GIS spatial analysis were used to explore the chlorophyll change and spatial distribution patterns of different varieties of cotton. By comparing with the spatial distribution pattern of Verticillium wilt, the correlation between chlorophyll and Verticillium wilt disease index was further analyzed, and the disease resistance characteristics of different varieties of cotton can be detected by measuring the chlorophyll content data of cotton canopy. That is, the correlation analysis of chlorophyll and Verticillium wilt through cluster analysis and spatial analysis was realized, and the resistance of different varieties of cotton to Verticillium wilt in different growth periods was detected.

[0088] Through cluster analysis and spatial analysis methods, the chlorophyll content data of different cotton varieties at multiple key growth stages were clustered. The analysis of the clustering results helped to find the significant differences in the distribution of the Verticillium wilt disease index of each variety and to build a multi-variety cotton chlorophyll Verticillium wilt detection model. Exemplarily, the optimized chlorophyll detection model was used for K-means cluster analysis to obtain five clusters with obvious differences. Then, through GIS spatial analysis, the chlorophyll change and spatial distribution patterns of different varieties of cotton were explored for further analysis of the association with the Verticillium wilt disease index.

[0089] As an example, the above process of performing K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton may specifically include the following steps: performing K-means cluster analysis based on the optimal chlorophyll detection model to obtain multiple clusters with different characteristics, and obtaining the chlorophyll change law and spatial distribution law of different varieties of cotton through GIS spatial analysis based on the multiple clusters.

[0090] In an optional embodiment, the process of determining the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of different varieties of cotton may specifically include the following steps: comparing the chlorophyll change law data and the chlorophyll spatial distribution law data with the spatial distribution law of Verticillium wilt of different varieties of cotton, and analyzing the correlation between chlorophyll and the disease index of Verticillium wilt of different varieties of cotton based on the comparison results, so as to detect the disease resistance characteristics of different varieties of cotton by measuring the chlorophyll content in the cotton canopy.

[0091] Step S180, determining the degree of association between chlorophyll in different growth stages and Verticillium wilt based on the correlation relationship, and based on the chlorophyll content data of different varieties of cotton measured in the field, determining the resistance level of different varieties of cotton to Verticillium wilt through the degree of association between chlorophyll and Verticillium wilt, and taking the resistance level of different varieties of cotton to Verticillium wilt as the result of cotton Verticillium wilt resistance detection.

[0092] In practical applications, such as Figure 2 As shown, based on the UAV chlorophyll detection model and analysis method, the resistance levels of different cotton varieties to Verticillium wilt were detected based on the correlation between chlorophyll and Verticillium wilt in different growth periods, thereby providing a reference for the prevention and control of Verticillium wilt at the field scale of cotton, which is conducive to increasing yield.

[0093] In the embodiments of the present application, before using chlorophyll to detect the resistance of multiple varieties of cotton to Verticillium wilt, drone data collection, chlorophyll measured data collection and Verticillium wilt disease index collection are first used. After the collected data are analyzed and processed, a drone chlorophyll detection model is constructed and optimized using a machine learning algorithm. Based on this model, the purpose of detecting the resistance level of multiple varieties of cotton to Verticillium wilt can be achieved by simply measuring the chlorophyll content data of the cotton canopy. Moreover, with the help of a high-throughput, high-precision drone remote sensing platform, the area and degree of Verticillium wilt occurrence in the field can be detected in a timely and accurate manner, and the resistance level of different cotton varieties to Verticillium wilt can be effectively detected, thereby improving the efficiency of cotton Verticillium wilt resistance detection and providing positive guidance and reference for cotton field disease resistance detection and breeding.

[0094] Through unmanned aerial vehicles and remote sensing technology, the detection and collection of phenotypic data of crops throughout their growth period can be significantly enhanced, covering multi-variety and time series data collection in field environments. In the early stage of the onset of Verticillium wilt, due to the degradation of the root cell wall of the crop by the secretion of toxins by Verticillium dahliae, after the cell wall is degraded, it colonizes in the xylem of the plant, blocks the xylem vessels, and destroys the normal physiological process. In the middle and late stages of the onset of Verticillium wilt, the infection of Verticillium dahliae by crops will cause significant changes in the microstructure of the leaves, resulting in a decrease in the photosynthetic activity of the mesophyll cells, causing the leaves to turn yellow, wither, and other damage. Therefore, chlorophyll fluorescence parameters and spectral characteristics are feasible in detecting Verticillium wilt in the asymptomatic period of cotton. In the embodiment of the present application, a dynamic detection model of chlorophyll for multi-variety cotton based on a specific infection experiment of Verticillium wilt is used for data processing, which can detect Verticillium wilt resistance for different cotton varieties based on the chlorophyll dynamic detection model on a field scale, thereby improving the efficiency of cotton Verticillium wilt resistance detection. Furthermore, the chlorophyll wilt detection model provides a low-cost, efficient, fast and accurate method to determine whether cotton within the planting range is infected with Verticillium wilt, so that the disease can be effectively detected and managed. By measuring the chlorophyll content data of the cotton canopy, the disease resistance characteristics of different varieties of cotton can be detected, which can also help users select disease-resistant varieties and increase cotton yields.

[0095] The method for detecting the resistance of multiple varieties of cotton to Verticillium wilt in the field based on the drone chlorophyll detection model provided in this solution helps users to measure the chlorophyll data of the cotton canopy in the field. After that, they only need to use the chlorophyll detection model built based on drone high-throughput data and machine learning algorithms provided in this solution to directly and automatically detect the resistance to Verticillium wilt of different varieties of cotton, as well as the identification of the area and degree of Verticillium wilt. This provides users with a highly efficient, high-precision, simple and practical method for detecting Verticillium wilt in cotton, thereby improving the efficiency of detecting Verticillium wilt resistance in cotton.

[0096] The above steps are described in detail below.

[0097] In some embodiments, after the above step S110, the method may further include the following steps: performing image preprocessing on the RGB digital orthophoto in the RGB data and the multispectral digital orthophoto in the multispectral data, wherein the image processing method of the image preprocessing includes at least one of the following: radiation calibration, geometric correction, image alignment, camera alignment optimization, dense point cloud generation, depth map creation, generation of RGB digital orthophoto model and generation of multispectral digital orthophoto model.

[0098] UAV data is used for chlorophyll detection model construction and optimization, which requires image preprocessing in advance, such as Figure 2As shown in the figure, the preprocessing process of the data collected by the drone includes: radiometric calibration, geometric correction, image alignment, camera optimization, dense point cloud generation, deep map creation and digital orthophoto model (DOM) generation, including the generation of RGB digital orthophoto model and multispectral digital orthophoto model. The above image preprocessing process can improve the efficiency of subsequent chlorophyll detection model construction and optimization.

[0099] Figure 3 The present invention provides a schematic diagram of the structure of a cotton Verticillium wilt resistance detection device based on a drone chlorophyll detection model. Figure 3 As shown, the cotton Verticillium wilt resistance detection device 300 based on the drone chlorophyll detection model includes:

[0100] The acquisition module 301 is used to acquire RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth periods of cotton in the field through image acquisition equipment arranged on the drone;

[0101] The collection module 302 is used to collect chlorophyll content data of different varieties of cotton measured in the designated infected field area and the designated uninfected field area respectively by using a chlorophyll analyzer; the time difference between the collection time corresponding to the drone and the collection time corresponding to the chlorophyll analyzer is less than a specified time period;

[0102] The calculation module 303 is used to calculate the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt severity disease classification standard; wherein the disease index is used to characterize the disease severity of Verticillium wilt;

[0103] An extraction module 304 is used to input the RGB digital orthophoto in the RGB data into an image automatic recognition and classification model through an image classification code, automatically perform image classification based on the RGB digital orthophoto and divide the cotton canopy range through the image automatic recognition and classification model, and extract cotton canopy vector boundaries of multiple varieties of cotton in different growth stages based on the image classification results and the cotton canopy range division results;

[0104] The analysis module 305 is used to calculate multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extract the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and perform a correlation analysis based on the mean of the vegetation indices and the disease index of the cotton Verticillium wilt, so as to obtain the optimal vegetation index sensitive to the cotton Verticillium wilt;

[0105] A construction module 306 is used to construct and optimize the chlorophyll detection model of different varieties of cotton at different growth stages by adjusting the initial model parameters based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average of the optimal vegetation index and the chlorophyll content data of different varieties of cotton measured on the spot, so as to obtain the optimal chlorophyll detection model;

[0106] The first determination module 307 is used to perform K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain chlorophyll variation law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll variation law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of the different varieties of cotton;

[0107] The second determination module 308 is used to determine the degree of association between chlorophyll in different growth periods and Verticillium wilt based on the association relationship, and based on the chlorophyll content data of the different varieties of cotton measured in the field, determine the resistance level of the different varieties of cotton to Verticillium wilt through the degree of association between chlorophyll and Verticillium wilt, and use the resistance level of the different varieties of cotton to Verticillium wilt as the cotton Verticillium wilt resistance detection result.

[0108] The cotton wilt resistance detection device based on the drone chlorophyll detection model provided in the embodiment of the present application has the same technical features as the cotton wilt resistance detection method based on the drone chlorophyll detection model provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0109] An electronic device provided in an embodiment of the present application is Figure 4 As shown, the electronic device 400 includes a processor 402 and a memory 401, wherein the memory stores a computer program that can be run on the processor, and the processor implements the steps of the method provided in the above embodiment when executing the computer program.

[0110] See also Figure 4The electronic device further includes: a bus 403 and a communication interface 404, a processor 402, a communication interface 404 and a memory 401 are connected via the bus 403; the processor 402 is used to execute an executable module stored in the memory 401, such as a computer program.

[0111] The memory 401 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 404 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0112] The bus 403 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0113] Among them, the memory 401 is used to store programs, and the processor 402 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the present application can be applied to the processor 402 or implemented by the processor 402.

[0114] The processor 402 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 402. The above processor 402 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to execute, or the hardware and software modules in the decoding processor are combined to execute. The software module may be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 401, and the processor 402 reads the information in the memory 401 and completes the steps of the above method in combination with its hardware.

[0115] Corresponding to the above-mentioned cotton Verticillium wilt resistance detection method based on the drone chlorophyll detection model, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to execute the steps of the above-mentioned cotton Verticillium wilt resistance detection method based on the drone chlorophyll detection model.

[0116] The cotton wilt resistance detection device based on the drone chlorophyll detection model provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in the embodiment of the present application are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment. Technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0117] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0118] For another example, the flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0121] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or in other words, the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the cotton wilt resistance detection method based on the unmanned aerial vehicle chlorophyll detection model described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0122] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0123] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solution of the present application, rather than to limit it. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solution recorded in the aforementioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiment of the present application. They should all be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for detecting cotton Verticillium wilt resistance based on an unmanned aerial vehicle chlorophyll detection model, characterized in that: The method comprises: The image acquisition equipment installed on the drone is used to obtain RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth stages of cotton in the field; Using a chlorophyll analyzer, field-measured chlorophyll content data of different varieties of cotton are collected in the designated infected field area and the designated uninfected field area; the time difference between the collection time corresponding to the drone and the collection time corresponding to the chlorophyll analyzer is less than a designated time period; Calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt severity disease grading standard; wherein the disease index is used to characterize the disease severity of Verticillium wilt; The RGB digital orthophoto in the RGB data is input into an image automatic recognition and classification model through an image classification code, and the image classification is automatically performed based on the RGB digital orthophoto and the cotton canopy range is divided through the image automatic recognition and classification model, and the cotton canopy vector boundaries of multiple varieties of cotton in different growth stages are extracted based on the image classification results and the cotton canopy range division results; Calculating multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extracting the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and performing a correlation analysis based on the mean of the vegetation indices and the disease index of cotton Verticillium wilt, to obtain the optimal vegetation index sensitive to cotton Verticillium wilt; Based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average of the optimal vegetation index, and the chlorophyll content data of different varieties of cotton measured on the spot, a chlorophyll detection model of different varieties of cotton chlorophyll at different growth stages is constructed and optimized by adjusting initial model parameters using a machine learning algorithm to obtain an optimal chlorophyll detection model; Based on the optimal chlorophyll detection model, K-means cluster analysis and GIS spatial analysis are performed to obtain chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of the different varieties of cotton; According to the correlation relationship, the degree of correlation between chlorophyll and Verticillium wilt in different growth periods is determined, and based on the chlorophyll content data of the different varieties of cotton measured in the field, the resistance levels of the different varieties of cotton to Verticillium wilt are determined through the degree of correlation between chlorophyll and Verticillium wilt, and the resistance levels of the different varieties of cotton to Verticillium wilt are used as the results of cotton Verticillium wilt resistance detection.

2. The method according to claim 1, characterized in that The disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease occurrence is calculated according to the designated Verticillium wilt disease severity classification standard, including: According to the designated cotton Verticillium wilt severity classification standard, the disease index of cotton Verticillium wilt of different varieties during the cotton Verticillium wilt outbreak period is calculated by the following formula: Among them, Z mi is the number of different cotton Verticillium wilt severity levels, n is the number of the highest cotton Verticillium wilt severity level, Z pi is the number of plants in each disease severity level of cotton Verticillium wilt, and i is the disease severity level of cotton Verticillium wilt.

3. The method according to claim 1, characterized in that After obtaining the RGB data and multispectral data of the designated infected field area and the designated uninfected field area at different key growth periods of the cotton field by the image acquisition device set on the drone, the method further includes: Image preprocessing is performed on the RGB digital orthophoto in the RGB data and the multispectral digital orthophoto in the multispectral data, wherein the image preprocessing includes at least one of the following: radiation calibration, geometric correction, image alignment, camera alignment optimization, dense point cloud generation, depth map creation, generation of the RGB digital orthophoto model, and generation of the multispectral digital orthophoto model.

4. The method according to claim 3, characterized in that The plurality of vegetation indices include at least one of the following: normalized difference vegetation index NDVI, difference vegetation index DVI, green-red vegetation index GRVI, infrared percentage vegetation index IPVI, and green normalized difference vegetation index GNDVI; the method further includes: Acquire a multispectral digital orthophoto after image preprocessing in the multispectral data; wherein the multispectral digital orthophoto after image preprocessing is an image synthesized by green light, red light, red edge and near infrared bands; The image synthesized by the green light, red light, red edge and near infrared bands is band-splitting based on the multiple vegetation indices sensitive to chlorophyll to obtain multiple single bands after band splitting, and the multiple single bands are used for calculation to obtain the optimal vegetation index sensitive to cotton Verticillium wilt.

5. The method according to claim 4, characterized in that The step of extracting the mean value of the optimal vegetation index corresponding to different varieties of cotton according to the cotton canopy vector boundary includes: The mean values ​​of the optimal vegetation index corresponding to different varieties of cotton are extracted using the QGIS program according to the cotton canopy vector boundary and the optimal vegetation index.

6. The method according to claim 1, characterized in that The method comprises: constructing and optimizing a chlorophyll detection model of different varieties of cotton at different growth stages by adjusting initial model parameters based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average value of the optimal vegetation index, and the chlorophyll content data of different varieties of cotton measured on the spot, and obtaining an optimal chlorophyll detection model, including: Based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index and the mean of the optimal vegetation index, three machine learning algorithms are used to construct and optimize the chlorophyll detection model of different varieties of cotton chlorophyll in different growth periods by adjusting the initial model parameters, and the accuracy of the chlorophyll detection model is detected by calculating the determination coefficient, root mean square error and mean absolute error to obtain the optimal chlorophyll detection model, wherein the higher the value of the determination coefficient and the lower the values ​​of the root mean square error and the mean absolute error, the higher the accuracy of the chlorophyll detection result corresponding to the chlorophyll detection model; the three machine learning algorithms include at least random forest, support vector machine and BP neural network.

7. The method according to claim 1, characterized in that The K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model are performed to obtain chlorophyll variation law data and chlorophyll spatial distribution law data of different varieties of cotton, including: K-means cluster analysis is performed based on the optimal chlorophyll detection model to obtain multiple clusters with different characteristics. Based on the multiple clusters, GIS spatial analysis is performed to obtain the chlorophyll change law and spatial distribution law of different varieties of cotton.

8. The method according to claim 1, characterized in that The determining of the correlation between the chlorophyll variation law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of different varieties of cotton includes: The chlorophyll change law data and the chlorophyll spatial distribution law data are compared with the spatial distribution laws of the incidence of Verticillium wilt in the different varieties of cotton, and the correlation between chlorophyll and the disease index of Verticillium wilt in the different varieties of cotton is analyzed based on the comparison results, so as to detect the disease resistance characteristics of different varieties of cotton by measuring the chlorophyll content in the cotton canopy.

9. A cotton Verticillium wilt resistance detection device based on an unmanned aerial vehicle chlorophyll detection model, characterized in that: include: An acquisition module is used to acquire RGB data and multispectral data of designated infected field areas and designated uninfected field areas at different key growth periods of cotton in the field through an image acquisition device set on the UAV; A collection module is used to collect chlorophyll content data of different varieties of cotton measured in the designated infected field area and the designated uninfected field area respectively by using a chlorophyll analyzer; the time difference between the collection time corresponding to the drone and the collection time corresponding to the chlorophyll analyzer is less than a specified time period; A calculation module, for calculating the disease index of Verticillium wilt of different varieties of cotton during the period of Verticillium wilt disease according to the designated cotton Verticillium wilt severity disease classification standard; wherein the disease index is used to characterize the disease severity of Verticillium wilt; An extraction module is used to input the RGB digital orthophoto in the RGB data into an image automatic recognition and classification model through an image classification code, automatically perform image classification based on the RGB digital orthophoto and divide the cotton canopy range through the image automatic recognition and classification model, and extract cotton canopy vector boundaries of multiple varieties of cotton in different growth stages based on the image classification results and the cotton canopy range division results; An analysis module, for calculating multiple vegetation indices sensitive to chlorophyll based on the multispectral digital orthoimage in the multispectral data, extracting the mean of the vegetation indices corresponding to different varieties of cotton according to the cotton canopy vector boundary, and performing a correlation analysis based on the mean of the vegetation indices and the disease index of the cotton Verticillium wilt, to obtain the optimal vegetation index sensitive to cotton Verticillium wilt; A construction module is used to construct and optimize the chlorophyll detection model of different varieties of cotton chlorophyll in different growth stages by adjusting the initial model parameters based on the RGB digital orthophoto, the multispectral digital orthophoto, the optimal vegetation index, the average of the optimal vegetation index and the chlorophyll content data of different varieties of cotton measured on the spot, so as to obtain the optimal chlorophyll detection model; The first determination module is used to perform K-means cluster analysis and GIS spatial analysis based on the optimal chlorophyll detection model to obtain chlorophyll change law data and chlorophyll spatial distribution law data of different varieties of cotton, and determine the correlation between the chlorophyll change law data, the chlorophyll spatial distribution law data and the disease index of Verticillium wilt of the different varieties of cotton; The second determination module is used to determine the degree of association between chlorophyll in different growth periods and Verticillium wilt according to the association relationship, and based on the chlorophyll content data of the different varieties of cotton measured in the field, determine the resistance level of different varieties of cotton to Verticillium wilt through the degree of association between chlorophyll and Verticillium wilt, and use the resistance level of different varieties of cotton to Verticillium wilt as the cotton Verticillium wilt resistance detection result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the method according to any one of claims 1 to 8.

Images