Cotton data processing method and device based on unmanned aerial vehicle and electronic equipment

Abstract

The invention provides a cotton data processing method and device based on an unmanned aerial vehicle and electronic equipment, relates to the technical field of biology, and solves the technical problem of relatively low data processing efficiency of cotton plant height genes. According to the method, RGB and a laser radar sensor are comprehensively utilized to obtain images and point cloud data in multiple growth periods of cotton, and through script batch identification and automatic calibration of cotton population canopy boundaries, multiple machine learning screening of an optimal model, plant height data and population genotype data association analysis, and the cotton population canopy boundary is identified. According to the method, canopy boundary calibration can be automatically completed in batches, two sensors are combined with multiple models to verify data more accurately, plant height related genes can be accurately identified in combination with phenotypic data in multiple periods, accuracy and efficiency are high, operation is simple and feasible, and the method is suitable for popularization and application. And the data processing efficiency of the cotton plant height gene is improved.

Description

Technical Field

[0001] The present application relates to the field of biotechnology, and in particular to a cotton data processing method, device and electronic device based on drone. Background Art

[0002] Cotton is one of the most important cash crops in the world and a major producer of natural fibers for the global textile industry. As a comprehensive agronomic trait, cotton plant type refers to the three-dimensional structure of the aboveground part of cotton. Cotton plant type is a key factor affecting cotton yield and mechanized harvesting efficiency. Improving cotton plant type can speed up the breeding process, effectively increase cotton yield, improve fiber quality, and thus improve the economic benefits of cotton planting. At present, the main cultivation characteristics of cotton are dwarfing, dense planting and early maturity, among which dwarfing refers to plant height.

[0003] Plant height is one of the most important factors in cotton plant type. Optimizing plant height can increase planting density, thereby improving the efficiency and yield of mechanized cotton harvesting. By introducing dwarfing genes into crops, plant height can be changed and yield can be significantly increased. However, improving plant height through traditional breeding methods such as hybrid breeding requires a lot of time in the field to select and cultivate hybrid combinations, and it is impossible to quickly and accurately improve the target material. By manually measuring field cotton plant height data and performing association analysis with genotype data, plant height genes can be identified and cotton materials can be improved, which can shorten the target material improvement process. However, the existing technology has low data processing efficiency for cotton plant height genes.

[0004] Application Example Content

[0005] The purpose of the embodiments of the present application is to provide a cotton data processing method, device and electronic device based on drone to solve the technical problem of low data processing efficiency of cotton plant height genes.

[0006] In a first aspect, the present application provides a cotton data processing method based on a drone, wherein the drone flies over a cotton field and is equipped with an RGB sensor and a lidar sensor; the method comprises:

[0007] Collecting RGB image data of the cotton field by the RGB sensor, and collecting laser point cloud data of the cotton field by the lidar sensor;

[0008] The RGB image data are aligned and spliced ​​to obtain a first processing result; the RGB image data are automatically recognized in batches by running a script according to the first processing result to obtain a first recognition result; a first soil part and a first background part in the RGB image data are divided according to the first recognition result, and all first pixels of the green vegetation part in the RGB image data are retained; a first group canopy range boundary is created for each material in the cotton group in the RGB image data according to the first pixel point, the first soil part and the first background part; a digital orthophoto map is constructed according to the first group canopy range boundary and a first digital surface model DSM is generated; and first plant height data is obtained based on the first digital surface model DSM;

[0009] Classifying and denoising the laser point cloud data to obtain a second processing result; dividing the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result; automatically identifying the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result; dividing the second soil part and the second background part in the laser point cloud data according to the second identification result, and retaining all second pixel points of the green vegetation part in the laser point cloud data; creating a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part; generating a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure-from-motion (SfM); generating a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM; and obtaining second plant height data based on the difference between the digital elevation model DEM and the second digital surface model DSM;

[0010] Acquire the plant height data of some individual plants in the cotton field measured in the field; wherein the plant height data measured in the field is obtained by a measurement process from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than a preset time period;

[0011] Using a plurality of different machine learning algorithms to respectively train and correct the first plant height data, the second plant height data, and the field-measured plant height data to obtain final first plant height data and final second plant height data;

[0012] Filtering the cotton population genome variation data SNPs in the cotton field to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than before filtering, calculating the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and performing GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtaining plant height gene data based on the analysis results;

[0013] Extracting the coding sequence of the located plant height-related gene based on the plant height gene data, amplifying the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, inserting the amplified fragment into a designated target vector to obtain a target gene overexpression vector, and generating gene overexpression new material data based on the target gene overexpression vector;

[0014] Based on the plant height gene data, the sgRNA sequence of the plant height-related gene is determined, the sgRNA sequence is inserted into a specified gene knockout vector to obtain a target gene knockout vector, and gene editing new material data is generated based on the target gene knockout vector.

[0015] In a possible implementation, the using of a plurality of different machine learning algorithms to respectively train and correct the first plant height data, the second plant height data, and the field measured plant height data to obtain the final first plant height data and the final second plant height data includes:

[0016] Using a variety of different machine learning algorithms through the determination coefficient R 2 , mean absolute error (MAE) and root mean square error (RMSE) to detect the extraction accuracy of the plant height data and obtain a comprehensive test result;

[0017] The best machine learning model result is selected according to the comprehensive detection result, and the first plant height data corresponding to the best machine learning model result is determined as the final first plant height data, and the second plant height data corresponding to the best machine learning model result is determined as the final second plant height data; wherein the final first plant height data and the final second plant height data are used in the GWAS analysis process.

[0018] In one possible implementation, the coefficient of determination R 2 The larger the MAE and the smaller the RMSE, the higher the accuracy of the model trained by the machine learning algorithm.

[0019] In one possible implementation, the transformation of the plant height-related genes utilizes the cotton efficient genetic transformation system SAMT without genotype restrictions. The cotton efficient genetic transformation system SAMT is used to use the cotton seed apical meristem stem cells as explants, combined with Agrobacterium and ultrasonic treatment, to integrate exogenous vectors into the stem cells to induce the stem cells to produce adventitious buds, and to use spectinomycin to inhibit the growth of the main buds to induce the production of axillary buds.

[0020] In a possible implementation, the script is used to divide the soil and the cotton plant canopy by color segmentation using different color spectra of the soil and the cotton plant canopy, and the boundary of the cotton plant canopy is automatically identified and calibrated based on the color spectrum.

[0021] In a possible implementation, obtaining first plant height data based on the first digital surface model DSM includes:

[0022] Based on the first digital surface model DSM, the difference between the upper percentile and the lower percentile is determined as the plant height value extracted from the RGB image data to obtain first plant height data.

[0023] In one possible implementation, the multiple different machine learning algorithms include any of the following:

[0024] Random Forest Regression, K-Nearest Neighbor, Huber Regression, Least Angle Regression, Lightweight Gradient Boosting Machine, Decision Tree Regression, Extra Trees Regression, Orthogonal Matching Pursuit, Bayesian Ridge Regression, AdaBoost, CatBoost, and Ridge Regression.

[0025] In a second aspect, the present application provides a cotton data processing device based on a drone, wherein the drone flies over a cotton field and is equipped with an RGB sensor and a laser radar sensor; the device comprises:

[0026] An acquisition module, used for acquiring RGB image data of the cotton field through the RGB sensor, and acquiring laser point cloud data of the cotton field through the lidar sensor;

[0027] A first processing module is used to align and splice the RGB image data to obtain a first processing result, automatically identify the RGB image data in batches by running a script according to the first processing result to obtain a first identification result, divide the RGB image data into a first soil part and a first background part according to the first identification result, and retain all first pixel points of the green vegetation part in the RGB image data, create a first group canopy range boundary for each material in the cotton group in the RGB image data according to the first pixel points, the first soil part and the first background part, and construct a digital orthophoto map according to the first group canopy range boundary and generate a first digital surface model DSM, and obtain first plant height data based on the first digital surface model DSM;

[0028] a second processing module, for classifying and denoising the laser point cloud data to obtain a second processing result, dividing the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result, automatically identifying the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result, dividing the laser point cloud data into a second soil part and a second background part according to the second identification result, and retaining all second pixel points of the green vegetation part in the laser point cloud data, creating a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part, and generating a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure from motion SfM, generating a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM, and obtaining second plant height data based on the digital elevation model DEM and the second digital surface model DSM;

[0029] An acquisition module, used for acquiring the field-measured plant height data of some individual plants in the cotton field; wherein the field-measured plant height data is obtained through a measurement process from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than a preset time period;

[0030] A training module, used to use a plurality of different machine learning algorithms to respectively train and verify the first plant height data, the second plant height data, and the field-measured plant height data to obtain final first plant height data and final second plant height data;

[0031] An analysis module is used for filtering the cotton population genome variation data SNPs in the cotton field to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than before filtering, calculating the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and performing GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtaining plant height gene data based on the analysis results;

[0032] The first construction module is used to extract the coding sequence of the located plant height-related gene based on the plant height gene data, amplify the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, and insert the amplified fragment into a specified target vector to obtain a target gene overexpression vector, and generate gene overexpression new material data based on the target gene overexpression vector;

[0033] The second construction module is used to determine the sgRNA sequence of the plant height-related gene based on the plant height gene data, insert the sgRNA sequence into the specified gene knockout vector to obtain the target gene knockout vector, and generate gene editing new material data based on the target gene knockout vector.

[0034] 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.

[0035] 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.

[0036] This application brings the following beneficial effects:

[0037] The present application provides a cotton data processing method, device and electronic device based on drone, which can collect RGB image data of cotton fields through the RGB sensor carried by the drone, collect laser point cloud data of cotton fields through the lidar sensor carried by the drone, align and splice the RGB image data to obtain a first processing result, automatically batch recognize the RGB image data by running a script according to the first processing result to obtain a first recognition result, divide the first soil part and the first background part in the RGB image data according to the first recognition result, retain all the first pixel points of the green vegetation part in the RGB image data, and align and splice the RGB image data according to the first pixel points, the first soil part and the first background part. For each material in the cotton population, a first population canopy range boundary is created, and a digital orthophoto map is constructed based on the first population canopy range boundary and a first digital surface model DSM is generated, and first plant height data is obtained based on the first digital surface model DSM, and the laser point cloud data is classified and denoised to obtain a second processing result, and the laser point cloud data is divided into ground point type point cloud data and non-ground point type point cloud data according to the second processing result, and the laser point cloud data is automatically identified in batches by running a script according to the second processing result to obtain a second identification result, and the laser point cloud data is divided into a second soil part and a second background part according to the second identification result, and all second pixel points of the green vegetation part in the laser point cloud data are retained, and according to the second The pixel points, the second soil part and the second background part are used to create a second group canopy range boundary for each material in the cotton group in the laser point cloud data, and the digital elevation model DEM is generated by the second group canopy range boundary through the structure motion SfM based on the ground point type point cloud data, and the second digital surface model DSM is generated by the second group canopy range boundary through SfM based on the non-ground point type point cloud data, and the second plant height data is obtained based on the difference between the digital elevation model DEM and the second digital surface model DSM, and the field measured plant height data of some individual plants in the cotton field are obtained, wherein the field measured plant height data is obtained through a measurement process from the ground to the top of the stem of the individual plant, and the measurement time of the measurement process is consistent with the RGB image data and the laser point The time difference between the cloud data collection time is less than the preset time period. A variety of different machine learning algorithms are used to train and correct the first plant height data, the second plant height data and the field measured plant height data, respectively, to obtain the final first plant height data and the final second plant height data. The cotton population genome variation data SNPs in the cotton field are filtered to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than those before filtering. The best linear unbiased estimator BLUE is calculated based on the final first plant height data and the final second plant height data, and GWAS analysis is performed on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain the analysis results, and the plant height gene data is obtained based on the analysis results.Based on the plant height gene data, the coding sequence of the plant height-related gene located was extracted, and the plant height gene was amplified using specific primers according to the coding sequence to obtain the amplified fragment, which was inserted into the specified target vector to obtain the target gene overexpression vector, and the gene overexpression new material data was generated based on the target gene overexpression vector. The sgRNA sequence of the plant height-related gene was determined based on the plant height gene data, and the sgRNA sequence was inserted into the specified gene knockout vector to obtain the target gene knockout vector, and the gene editing new material data was generated based on the target gene knockout vector. In this scheme, RGB and lidar sensors were used to obtain images and point cloud data of cotton in multiple growth periods, and the cotton population canopy boundary was batch identified and automatically calibrated through scripts. The most suitable model was selected by multiple machine learning, and the plant height data was associated with the population genotype data. Gene overexpression and gene editing materials were obtained for verification, and the canopy boundary calibration was realized to be completed automatically in batches. The two sensors combined with multi-model verification data were more accurate, and the plant height-related genes could be accurately identified by combining multi-period phenotypic data. The accuracy and efficiency were high, and the operation was simple and feasible, which improved the data processing efficiency of cotton plant height genes.

[0038] 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

[0039] 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.

[0040] Figure 1 A schematic diagram of a cotton data processing method based on a drone provided in an embodiment of the present application;

[0041] Figure 2 Another schematic diagram of a cotton data processing method based on a drone provided in an embodiment of the present application;

[0042] Figure 3 An example of material planting and drone platform selection in the drone-based cotton data processing method provided in the embodiment of the present application;

[0043] Figure 4 An example of the automatic calibration result of the cotton canopy obtained in the cotton data processing method based on drone provided in the embodiment of the present application;

[0044] Figure 5An example of the results obtained by using machine learning on RGB and LiDAR data in the UAV-based cotton data processing method provided in the embodiment of the present application;

[0045] Figure 6 An example of the results obtained by performing GWAS on RGB and LiDAR data and genotype data in the UAV-based cotton data processing method provided in the embodiment of the present application;

[0046] Figure 7 An example of the results obtained by overexpressing and gene editing the GhPH_UAV1 gene in the cotton data processing method based on drones provided in the embodiments of the present application;

[0047] Figure 8 A schematic diagram of the structure of a cotton data processing device based on a drone provided in an embodiment of the present application;

[0048] Fig. 9 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] 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.

[0050] 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.

[0051] At present, the rapid development of genomics has improved the efficiency of crop breeding. By conducting genome-wide association studies (GWAS) on plant height data from multiple periods and genotype data, the genes that control plant height can be identified and new materials for plant height improvement can be quickly and accurately obtained through gene editing and other means, shortening the breeding process. Since plant height changes rapidly during cotton growth, it is crucial to quickly obtain plant height data from a large number of populations for GWAS analysis. Traditional field plant height data collection relies on manual measurement, which has significant disadvantages such as low throughput, high cost and strong subjectivity. Obtaining phenotypic data through indoor phenotyping platforms has improved throughput to a certain extent, but the high cost of these facilities and the inability to simulate field growth environments have greatly limited their application. High-throughput plant height phenotypic data acquisition and automated batch analysis in the field are still a technical bottleneck.

[0052] The rapid development of drone technology and imaging analysis technology has opened up new avenues for high-throughput measurement of crop field phenotypes and provided accurate phenotypic data for GWAS analysis. Compared with traditional phenotypic analysis methods, high-throughput phenotypic analysis can dynamically observe the traits of large sample populations at different growth stages and provide more objective trait representation based on images or point cloud data. However, the acquisition and analysis of field cotton plant height data using drone platforms still face the following bottlenecks: (1) The canopy of cotton populations changes in images at different stages, and the efficiency of conventional manual calibration is extremely low, making it impossible to achieve automated and accurate boundary calibration; (2) The plant height data obtained by a single sensor is one-sided and cannot be mutually verified by different data sets; (3) When training the plant height obtained by drones and the plant height measured manually, a comprehensive machine learning model should be selected for comparison to obtain the most suitable model data; (4) The current methods do not overexpress and gene edit the genes located by GWAS using drone plant height data, and successfully obtain cotton plant height gene overexpression and gene editing materials. This shows that the existing technology has low efficiency in processing data on cotton plant height genes and is currently unable to quickly and accurately achieve improvement of target materials.

[0053] Based on this, the embodiments of the present application provide a cotton data processing method, device and electronic device based on drone, through which the technical problem of low data processing efficiency of cotton plant height genes can be solved.

[0054] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0055] Figure 1 The following is a flow chart of a cotton data processing method based on drones provided in an embodiment of the present application. The drones fly over the cotton fields and are equipped with RGB sensors and lidar sensors. Figure 1 As shown, the method includes:

[0056] Step S110, collecting RGB image data of the cotton field through an RGB sensor, and collecting laser point cloud data of the cotton field through a lidar sensor.

[0057] As an optional implementation, Figure 2 As shown in the figure, drones equipped with RGB sensors and lidar sensors can be used to fly over cotton fields during a specific growth period of cotton to collect high-definition image data.

[0058] For example, the distribution of 419 materials in the experimental field (cotton field) is shown in Figure 3 A. The drone platforms used for image acquisition are Mavic 3E equipped with RGB sensor and Matrice 350RTK equipped with LiDAR sensor. Both drones are equipped with advanced integrated real-time kinematic (RTK) GNSS modules, which can provide positioning data with a positioning accuracy of 1 cm + 1 ppm (horizontally) and 1.5 cm + 1 ppm (vertically). Figure 3 B). The drone flights were conducted on June 11, June 18, June 26, July 3, and July 8, 2023, during which the cotton plants were in a rapid growth stage. On the day of the flight, some plant individuals were selected and manually measured from the ground to the top of the stem using a ruler to obtain the manually measured plant height data.

[0059] Step S120, aligning and splicing the RGB image data to obtain a first processing result, automatically identifying the RGB image data in batches by running a script according to the first processing result to obtain a first identification result, dividing the RGB image data into a first soil part and a first background part according to the first identification result, and retaining all first pixel points of the green vegetation part in the RGB image data, creating a first group canopy range boundary for each material in the cotton group in the RGB image data according to the first pixel point, the first soil part and the first background part, and constructing a digital orthophoto map according to the first group canopy range boundary and generating a first digital surface model DSM, and obtaining first plant height data based on the first digital surface model DSM.

[0060] For example, 419 cotton materials formed 419 different canopy structures in each growth period. Conventional methods of manual calibration are not only inefficient, but also unable to achieve accurate calibration of the canopy edge. The code can automatically recognize images in batches, and automatically identify and segment them according to the differences in color and space between the soil and the canopy, so as to create precise boundaries for each material in the cotton group, while minimizing the impact of the soil and maximizing the retention of the group canopy data. The effect is as follows Figure 4 shown.

[0061] like Figure 2As shown in the figure, the RGB photos of the five growth stages were aligned and spliced. Furthermore, the software DJI Terra V4.2.2 was used to construct a digital orthophoto map (DOM) and generate a digital surface model (DSM). The difference between the first percentile and the 95th percentile was selected as the plant height value extracted from the drone RGB image, and finally the plant height data of 419 materials in the five growth stages were obtained.

[0062] As an optional implementation, the script can be used to divide the soil and cotton group canopy by color segmentation using the different color spectra of the soil and the cotton group canopy, and the boundary of the cotton group canopy is automatically identified and calibrated based on the color spectrum. This script code uses the different color spectra of the soil and the cotton canopy to distinguish the soil and the canopy through color segmentation, so the boundary of the cotton group canopy can be automatically and accurately identified and calibrated based on the color. This step can support the subsequent RGB and LiDAR data to calculate the plant height of all individual plants of each material, thereby obtaining accurate cotton plant height data.

[0063] Running the script "Boundary demarcation.py" through PYTHON V3.8 can automatically identify images in batches, complete the conversion from color images to grayscale images, separate the soil and background, and retain all pixels of green vegetation, thereby creating precise boundaries for each material in the cotton group, minimizing the impact of soil while maximizing the retention of group canopy range data.

[0064] Furthermore, the script "Boundary demarcation.py" described in the embodiment of the present application is specifically:

[0065] import cv2

[0066] import numpy as np

[0067] Boundary1=cv2.imread('Cotton.tiff')

[0068] Boundary2=np.array(Boundary1,dtype=np.float32) / 255.0

[0069] (b,g,r)=cv2.split(Boundary2)

[0070] Gray=2.4*gbr

[0071] (minVal,maxVal,minLoc,maxLoc)=cv2.minMaxLoc(Gray)

[0072] Gray_u8=np.array((gray-minVal) / (maxVal-minVal)*255,dtype=np.uint8)

[0073] (thresh,th)=cv2.threshold(Gray_u8,-1.0,255,CV2.THRESH OTSU)

[0074] (b8,g8,r8)=cv2.split(Boundary1)

[0075] Color_pictures=cv2.merge([b8&th,g8&th,r8&th])

[0076] cv2.imshow("Boundary1",Color_pictures)

[0077] It should be noted that the code Boundary demarcation.py above creates precise boundaries for each material in the cotton population based on the following programming method: 1. Import the required libraries: cv2 (OpenCV library) and numpy; 2. Use the `imread` function of the OpenCV library to read the image file 'RGB.tiff' taken with visible light; 3. Convert the read image data into a NumPy array and normalize it to between 0 and 1 for further processing; 4. Use the `split` function of the OpenCV library to decompose the image into three color channels: B (blue), G (green), and R (red); 5. Calculate the grayscale image. A specific formula is used here: `gray=2.4*gbr`, which is to multiply the value of the green channel by 2.4 and then subtract the values ​​of the blue and red channels to obtain a grayscale image; 6. Use the `minMaxLoc` function of the OpenCV library to find the minimum and maximum values ​​in the grayscale image, as well as their locations; 7. Normalize the grayscale image to the range of 0-255 for threshold processing. A specific formula is used here: `(gray-minVal) / (maxVal-minVal)*255`, which is to subtract the minimum value (minVal) from the value of the grayscale (gray) image, then divide it by the difference between the maximum value (maxVal) and the minimum value (minVal), and finally multiply it by 255 to get a grayscale image in the range of 0-255; 8. Use the `threshold` function of the OpenCV library to perform threshold processing on the grayscale image. The Otsu algorithm is used here to automatically calculate the threshold; 9. Process the B, G, and R channels of the color image with the grayscale image after threshold processing to obtain a new color image; 10. Use the `merge` function of the OpenCV library to merge the new color image into one image; 12. Use the `imshow` function of the OpenCV library to display the processed image.

[0078] In a possible implementation, the software DJI Terra V4.2.2 may be used to align and stitch a large number of RGB photos, construct a digital orthophoto map, and generate a digital surface model (DSM).

[0079] The above process of obtaining the first plant height data based on the first digital surface model DSM may specifically include the following steps: based on the first digital surface model DSM, the difference between the upper percentile and the lower percentile is determined as the plant height value extracted from the RGB image data to obtain the first plant height data.

[0080] Considering that soil conditions vary with human activities and rainfall, the difference between the upper percentile and the lower percentile is selected as the plant height value extracted from the drone RGB image. Usually, the minimum first and second percentiles and the maximum 95-98 percentiles are selected as the bottom and top heights of the main stem on the ground, respectively. In the embodiment of the present application, the difference between the second percentile and the 98th percentile is selected as the plant height value extracted from the drone RGB image.

[0081] Step S130, classifying and denoising the laser point cloud data to obtain a second processing result, dividing the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result, automatically identifying the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result, dividing the second soil part and the second background part in the laser point cloud data according to the second identification result, and retaining all second pixel points of the green vegetation part in the laser point cloud data, creating a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part, and generating a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure from motion SfM, generating a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM, and obtaining second plant height data based on the difference between the digital elevation model DEM and the second digital surface model DSM.

[0082] For example, the software DJI Terra V4.2.2 is used to roughly classify the acquired laser point cloud data into two categories: ground points and non-ground points, such as Figure 2 As shown in the figure, the point cloud data is denoised to reduce the noise in the point cloud. The ground point cloud data is converted into a digital elevation model (DEM) through the structure-from-motion (SfM) technology, and the non-ground point cloud data is converted into a digital surface model (DSM). For example, the plant height data of 419 materials in 5 periods are calculated by the difference between DEM and DSM.

[0083] As a possible implementation method, the cotton plant height obtained by LiDAR is calculated by the difference between DSM and DEM.

[0084] Step S140, obtaining plant height data of some individual plants in the cotton field measured on site.

[0085] The plant height data measured on site is obtained by measuring from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than the preset time period. Figure 2 As shown, on the day of the drone flight, some plant individuals are selected and manually measured from the ground to the top of the stem using a ruler to obtain manually measured plant height data. For another example, the plant height measurement process is performed immediately after the drone is collected, so that the plant height will not change in a short time, ensuring the consistency of the data.

[0086] Step S150, using a plurality of different machine learning algorithms to respectively train and calibrate the first plant height data, the second plant height data, and the field measured plant height data to obtain final first plant height data and final second plant height data.

[0087] Among them, the multiple different machine learning algorithms can include any of the following: random forest regression algorithm, K-nearest neighbor algorithm, Huber regression algorithm, least angle regression algorithm, lightweight gradient boosting machine algorithm, decision tree regression algorithm, extra tree regression algorithm, orthogonal matching pursuit algorithm, Bayesian ridge regression algorithm, AdaBoost regression algorithm, CatBoost regression algorithm, and ridge regression algorithm.

[0088] In practical applications, different machine learning methods have different effects. For example, Figure 2As shown in the figure, the following 12 machine learning algorithms were used: Random Forest Regression (RFR), K-Nearest Neighbors (KNN), Huber Regression, Least Angle Regression, Light Gradient Boosting Machine (LightGBM), Decision Tree Regressor, Extra Trees Regressor, Orthogonal Matching Pursuit, Bayesian Ridge Regressor, AdaBoost Regressor, CatBoost Regressor and Ridge Regressor to train and verify the plant height values ​​extracted by RGB and LiDAR and the plant height values ​​measured manually. The extraction accuracy of plant height was evaluated using three indicators: determination coefficient (R 2 ), mean absolute error (MAE), and root mean square error (RMSE). The best model result was selected as the final plant height data based on the comprehensive results for subsequent GWAS analysis.

[0089] As an optional implementation, a variety of different machine learning algorithms are used to determine the coefficient R 2 , mean absolute error (MAE), and root mean square error (RMSE) are used to test the extraction accuracy of plant height data to obtain a comprehensive test result; the best machine learning model result is selected according to the comprehensive test result, and the first plant height data corresponding to the best machine learning model result is determined as the final first plant height data, and the second plant height data corresponding to the best machine learning model result is determined as the final second plant height data; wherein, the final first plant height data and the final second plant height data are used in the GWAS analysis process.

[0090] Among them, the coefficient of determination R 2 The larger the MAE and the smaller the RMSE, the higher the accuracy of the model trained by the machine learning algorithm. The plant height data obtained by RGB and LiDAR and the manual measurement data were trained and corrected using 12 machine learning algorithms to evaluate the accuracy of plant height prediction and finally generate a set of accuracy test results. 2) is larger, the mean absolute error (MAE) and root mean square error (RMSE) are smaller, and the model accuracy is better.

[0091] For example, 14 machine learning algorithms were used to train and validate a total of 2095 RGB and LiDAR plant height values ​​and manually measured plant height values ​​in five periods. In the model construction of RGB data, it was found that the model constructed by the least angle regression algorithm had the highest accuracy, the largest R2 (0.914), the smallest MAE (0.042) and the smallest RMSE (0.055). Figure 5 A). In the model construction of lidar data, it was found that the model constructed by the Huber regression algorithm had the highest accuracy, the largest R2 (0.934), and the smallest MAE (0.040) and RMSE (0.050) ( Figure 5 B). The final first plant height data and final second plant height data obtained by the two best models will be used for subsequent GWAS analysis and verification.

[0092] Step S160, filtering the cotton population genome variation data SNPs in the cotton field to obtain data whose actual quality and accuracy are greater than the target cotton population genome variation data SNPs before filtering, calculating the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and performing GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtaining plant height gene data based on the analysis results.

[0093] It should be noted that the process of calculating the best linear unbiased estimator BLUE is a common step in GWAS analysis, which is to convert the phenotypic data into this form before analyzing it together with the genotype.

[0094] As an optional implementation, the screening conditions of SNPs are set to MAF>0.05 and missing rate<10%. The significance threshold of GWAS is set to -log10(p)=5. In this step, the cotton population genomic variation data (SNPs) are filtered to obtain high-quality SNPs (MAF>0.05, missing rate<10%). The best linear unbiased estimator (BLUE) is calculated for the cotton population plant height data obtained by the drone platform, and GWAS analysis is performed with SNPs. The significance threshold of the association is set to -log10(p)=5, and plant height-related genes are obtained in the significant interval.

[0095] For example, Figure 2As shown in Figure 2, GWAS analysis was performed on all five time points of 419 cotton materials using lidar-based plant height data, and it was found that five peaks on chromosomes A01 (~3.75~3.88Mb), A02 (~43.8~44.5Mb), A07 (~27.8~28.2Mb), A10 (~110.7~111.1Mb) and D11 (~10.8~11.0Mb) were significantly correlated with plant height ( Figure 6 A). The same result was obtained based on the RGB analysis ( Figure 6 B). The two genes located on chromosome A01 are consistent with the cotton plant height gene GhUBP15 (Ye et al., 2023) identified using drone RGB phenotype data and the cotton plant height gene GhPH1 (Tian et al., 2024) identified using manual measurement data, indicating that the results obtained by the method of the embodiment of the present application are comprehensive and reliable. A new gene was identified from chromosome D11 that may be related to cotton plant height and was named GhPH_UAV1.

[0096] Step S170, extract the coding sequence of the located plant height-related gene based on the plant height gene data, amplify the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, insert the amplified fragment into a specified target vector to obtain a target gene overexpression vector, and generate gene overexpression new material data based on the target gene overexpression vector.

[0097] For example, the coding sequence of the gene related to plant height located is extracted, and specific primers are designed to amplify the gene. The amplified fragment is inserted into the pCAMBIA-2300 vector driven by the 35S promoter to construct an overexpression vector, such as Figure 2 As shown, a new material for gene overexpression is created. For example, the coding sequence of the GhPH_UAV1 gene is extracted, and specific primers are designed to amplify the gene. The primer information is shown in Table 1 below, i.e., the primer information of the GhPH_UAV1 gene:

[0098]

[0099] In addition, the transformation of plant height-related genes all used the cotton efficient genetic transformation system SAMT without genotype restrictions. The cotton efficient genetic transformation system SAMT is used to use cotton seed apical meristem stem cells as explants, combined with Agrobacterium and ultrasonic treatment, to integrate exogenous vectors into stem cells to induce stem cells to produce adventitious buds, and to use spectinomycin to inhibit the growth of the main buds to induce the production of axillary buds. Specifically, the transformation of plant height genes all used the cotton efficient genetic transformation system (SAMT) without genotype restrictions. Using cotton seed apical meristem stem cells as explants, combined with Agrobacterium and ultrasonic treatment, to integrate exogenous vectors into stem cells, thereby inducing stem cells to produce adventitious buds, and to use spectinomycin to inhibit the growth of the main buds and induce the production of axillary buds, effectively reducing the probability of chimeras. The obtained overexpression transgenic materials and gene-edited mutant materials can be stably inherited, and there is no sterility of transgenic regenerated seedlings, and no deformed seedlings.

[0100] Step S180, determine the sgRNA sequence of the plant height-related gene based on the plant height gene data, insert the sgRNA sequence into the specified gene knockout vector to obtain the target gene knockout vector, and generate gene editing new material data based on the target gene knockout vector.

[0101] As a possible implementation method, sgRNAs for plant height-related genes can be generated, and the sgRNA sequences can be inserted into CRISPR / Cas9 gene knockout vectors developed specifically for upland cotton to construct gene knockout vectors, such as Figure 2 As shown, creating new materials for gene editing.

[0102] In order to further clarify the role of GhPH_UAV1 in cotton PH, the present application example used CRISPR / Cas9 technology to create cotton materials with GhPH_UAV1 gene knockout (GhPH_UAV1-Cas9) and cotton materials with GhPH_UAV1 gene overexpression (GhPH_UAV1-OE). qRT-PCR analysis showed that compared with the cotton material ZM24 with normal plant height, the expression level of GhPH_UAV1 gene in GhPH_UAV1-OE material was significantly increased, proving that the GhPH_UAV1 gene overexpression experiment was successfully implemented ( Figure 7 A). Furthermore, compared with ZM24, the plant height of GhPH_UAV1-OE material increased by 16.48%, while the plant height of GhPH_UAV1-Cas9 material decreased by 21.44% ( Figure 7 BC). Furthermore, the present invention observed the cell development of the two materials. Compared with ZM24, the cell length of the GhPH_UAV1-OE material increased significantly, while the cell length of the GhPH_UAV1-Cas9 material decreased significantly ( Figure 7DE). The above results intuitively prove that the GhPH_UAV1 gene identified by the method of the present application example has a significant effect of positively regulating cotton plant height.

[0103] In the embodiments of the present application, RGB and lidar sensors are used in combination to obtain images and point cloud data of cotton at multiple growth periods, batch identification and automatic calibration of cotton population canopy boundaries are performed through scripts, multiple machine learning methods are used to screen the most suitable model, correlation analysis of plant height data and population genotype data is performed, and gene overexpression and gene editing materials are obtained for verification, so that canopy boundary calibration can be completed automatically in batches. The two sensors combined with multi-model verification data are more accurate, and plant height-related genes can be accurately identified in combination with phenotypic data from multiple periods, with high accuracy and efficiency, simple and feasible operation, thereby improving the data processing efficiency of cotton plant height genes.

[0104] By using two sensors to obtain plant height in the field, it can be used for comparison and verification of subsequent GWAS analysis, making phenotypic data more reliable; moreover, canopy boundary calibration can be completed automatically in batches through code, with high efficiency; furthermore, 14 machine learning methods are used for modeling, and the most suitable model is selected to make phenotypic data more accurate; in addition, the present application embodiment can perfectly and smoothly combine the acquisition and analysis of plant height data of large-scale groups in the field during multiple growth periods with GWAS identification of genes and verification of gene functions, so as to achieve accurate connection of phenotypic and genotypic data, and largely make up for the time-consuming and laborious manual investigation of phenotypic traits in the field, the subjectivity, and the lack of process and batch analysis during data analysis. The present application embodiment can play an important role in the process of acquiring and analyzing phenotypic traits of large-scale groups in the field during multiple growth periods.

[0105] The method provided in the embodiment of the present application can be used as a method for unmanned aerial vehicle multi-sensor measurement to reveal the genetic basis of cotton plant height. This method helps scientific researchers to no longer need to carry out the arduous work of manually measuring the plant height of large groups planted in the field. They only need to operate the drone to collect images during a specific period, and then batch processing can be achieved through the constructed plant height model. The plant height data of all population samples can be accurately obtained, which can be directly used for association analysis with genotype data to obtain genes that regulate plant height, and create new materials for targeted improvement of plant height through gene editing, thereby providing accurate and valuable gene and material resources for modern molecular breeding.

[0106] It should be noted that the experimental methods used in the examples of this application are conventional methods unless otherwise specified. The materials, reagents, etc. used in the examples of this application can be obtained from commercial channels unless otherwise specified.

[0107] Figure 8A schematic diagram of a cotton data processing device based on an unmanned aerial vehicle is provided. The unmanned aerial vehicle flies over a cotton field and is equipped with an RGB sensor and a laser radar sensor. Figure 8 As shown, the cotton data processing device 800 based on drone includes:

[0108] The acquisition module 801 is used to acquire RGB image data of the cotton field through the RGB sensor, and to acquire laser point cloud data of the cotton field through the laser radar sensor;

[0109] The first processing module 802 is used to align and splice the RGB image data to obtain a first processing result, automatically identify the RGB image data in batches by running a script according to the first processing result to obtain a first identification result, divide the first soil part and the first background part in the RGB image data according to the first identification result, and retain all first pixel points of the green vegetation part in the RGB image data, create a first group canopy range boundary for each material in the cotton group in the RGB image data according to the first pixel points, the first soil part and the first background part, and construct a digital orthophoto map according to the first group canopy range boundary and generate a first digital surface model DSM, and obtain first plant height data based on the first digital surface model DSM;

[0110] The second processing module 803 is used to classify and denoise the laser point cloud data to obtain a second processing result, divide the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result, automatically identify the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result, divide the second soil part and the second background part in the laser point cloud data according to the second identification result, and retain all second pixel points of the green vegetation part in the laser point cloud data, create a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part, and generate a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure from motion SfM, generate a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM, and obtain second plant height data based on the digital elevation model DEM and the second digital surface model DSM;

[0111] The acquisition module 804 is used to acquire the field-measured plant height data of some individual plants in the cotton field; wherein the field-measured plant height data is obtained through a measurement process from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than a preset time period;

[0112] A training module 805 is used to use a plurality of different machine learning algorithms to respectively train and verify the first plant height data, the second plant height data, and the field-measured plant height data to obtain final first plant height data and final second plant height data;

[0113] An analysis module 806 is used to filter the cotton population genome variation data SNPs in the cotton field to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than that before filtering, calculate the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and perform GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtain plant height gene data based on the analysis results;

[0114] The first construction module 807 is used to extract the coding sequence of the located plant height-related gene based on the plant height gene data, amplify the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, and insert the amplified fragment into a specified target vector to obtain a target gene overexpression vector, and generate gene overexpression new material data based on the target gene overexpression vector;

[0115] The second construction module 808 is used to determine the sgRNA sequence of the plant height-related gene based on the plant height gene data, insert the sgRNA sequence into the specified gene knockout vector to obtain the target gene knockout vector, and generate gene-edited new material data based on the target gene knockout vector.

[0116] The drone-based cotton data processing device provided in the embodiment of the present application has the same technical features as the drone-based cotton data processing method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0117] An electronic device provided in an embodiment of the present application is Fig. 9 As shown, the electronic device 900 includes a processor 902 and a memory 901, 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.

[0118] See also Fig. 9 The electronic device further includes: a bus 903 and a communication interface 904, a processor 902, a communication interface 904 and a memory 901 are connected via the bus 903; the processor 902 is used to execute an executable module stored in the memory 901, such as a computer program.

[0119] The memory 901 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 904 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0120] The bus 903 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, Fig. 9 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.

[0121] Among them, the memory 901 is used to store programs, and the processor 902 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 902 or implemented by the processor 902.

[0122] The processor 902 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 902. The above processor 902 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 901, and the processor 902 reads the information in the memory 901 and completes the steps of the above method in combination with its hardware.

[0123] Corresponding to the above-mentioned drone-based cotton data processing method, 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 drone-based cotton data processing method.

[0124] The cotton data processing device based on drone provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The device provided in the embodiment of the present application has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brief description, the parts not mentioned in the device embodiment can refer to the corresponding contents in the aforementioned method embodiment. Those skilled in the art 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.

[0125] 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.

[0126] 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.

[0127] 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.

[0128] 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.

[0129] 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 essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for 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 data processing method based on drones described in each embodiment of the present application. The aforementioned storage media include: 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.

[0130] 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.

[0131] 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 cotton data processing method based on drone, characterized in that: The drone flies over a cotton field, and the drone is equipped with an RGB sensor and a laser radar sensor; the method includes: Collecting RGB image data of the cotton field by the RGB sensor, and collecting laser point cloud data of the cotton field by the lidar sensor; The RGB image data are aligned and spliced ​​to obtain a first processing result; the RGB image data are automatically recognized in batches by running a script according to the first processing result to obtain a first recognition result; a first soil part and a first background part in the RGB image data are divided according to the first recognition result, and all first pixels of the green vegetation part in the RGB image data are retained; a first group canopy range boundary is created for each material in the cotton group in the RGB image data according to the first pixel point, the first soil part and the first background part; a digital orthophoto map is constructed according to the first group canopy range boundary and a first digital surface model DSM is generated; and first plant height data is obtained based on the first digital surface model DSM; Classifying and denoising the laser point cloud data to obtain a second processing result; dividing the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result; automatically identifying the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result; dividing the second soil part and the second background part in the laser point cloud data according to the second identification result, and retaining all second pixel points of the green vegetation part in the laser point cloud data; creating a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part; generating a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure-from-motion (SfM); generating a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM; and obtaining second plant height data based on the difference between the digital elevation model DEM and the second digital surface model DSM; Acquire the plant height data of some individual plants in the cotton field measured in the field; wherein the plant height data measured in the field is obtained by a measurement process from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than a preset time period; Using a plurality of different machine learning algorithms to respectively train and correct the first plant height data, the second plant height data, and the field-measured plant height data to obtain final first plant height data and final second plant height data; Filtering the cotton population genome variation data SNPs in the cotton field to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than before filtering, calculating the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and performing GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtaining plant height gene data based on the analysis results; Extracting the coding sequence of the located plant height-related gene based on the plant height gene data, amplifying the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, inserting the amplified fragment into a designated target vector to obtain a target gene overexpression vector, and generating gene overexpression new material data based on the target gene overexpression vector; Based on the plant height gene data, the sgRNA sequence of the plant height-related gene is determined, the sgRNA sequence is inserted into a specified gene knockout vector to obtain a target gene knockout vector, and gene editing new material data is generated based on the target gene knockout vector.

2. The method according to claim 1, characterized in that The method of using a plurality of different machine learning algorithms to respectively train and correct the first plant height data, the second plant height data, and the field measured plant height data to obtain the final first plant height data and the final second plant height data includes: Using a variety of different machine learning algorithms through the determination coefficient R 2 , mean absolute error (MAE) and root mean square error (RMSE) to detect the extraction accuracy of the plant height data and obtain a comprehensive test result; The best machine learning model result is selected according to the comprehensive detection result, and the first plant height data corresponding to the best machine learning model result is determined as the final first plant height data, and the second plant height data corresponding to the best machine learning model result is determined as the final second plant height data; wherein the final first plant height data and the final second plant height data are used in the GWAS analysis process.

3. The method according to claim 2, characterized in that The coefficient of determination R 2 The larger the MAE and the smaller the RMSE, the higher the accuracy of the model trained by the machine learning algorithm.

4. The method according to claim 1, characterized in that: The transformation of the plant height-related genes all utilizes the cotton efficient genetic transformation system SAMT without genotype restriction. The cotton efficient genetic transformation system SAMT is used to use cotton seed apical meristem stem cells as explants, combine Agrobacterium and ultrasonic treatment, integrate exogenous vectors into stem cells to induce stem cells to produce adventitious buds, and use spectinomycin to inhibit the growth of main buds to induce the production of axillary buds.

5. The method according to claim 1, characterized in that The script is used to divide the soil and the cotton colony canopy by color segmentation using the different color spectra of the soil and the cotton colony canopy. The boundary of the cotton colony canopy is automatically identified and calibrated based on the color spectrum.

6. The method according to claim 1, characterized in that The obtaining first plant height data based on the first digital surface model DSM includes: Based on the first digital surface model DSM, the difference between the upper percentile and the lower percentile is determined as the plant height value extracted from the RGB image data to obtain first plant height data.

7. The method according to claim 1, characterized in that The multiple different machine learning algorithms include any of the following: Random Forest Regression, K-Nearest Neighbor, Huber Regression, Least Angle Regression, Lightweight Gradient Boosting Machine, Decision Tree Regression, Extra Trees Regression, Orthogonal Matching Pursuit, Bayesian Ridge Regression, AdaBoost, CatBoost, and Ridge Regression.

8. A cotton data processing device based on drone, characterized in that: The drone flies over the cotton field, and the drone is equipped with an RGB sensor and a laser radar sensor; the device includes: An acquisition module, used for acquiring RGB image data of the cotton field through the RGB sensor, and acquiring laser point cloud data of the cotton field through the lidar sensor; A first processing module is used to align and splice the RGB image data to obtain a first processing result, automatically identify the RGB image data in batches by running a script according to the first processing result to obtain a first identification result, divide the RGB image data into a first soil part and a first background part according to the first identification result, and retain all first pixel points of the green vegetation part in the RGB image data, create a first group canopy range boundary for each material in the cotton group in the RGB image data according to the first pixel points, the first soil part and the first background part, and construct a digital orthophoto map according to the first group canopy range boundary and generate a first digital surface model DSM, and obtain first plant height data based on the first digital surface model DSM; a second processing module, for classifying and denoising the laser point cloud data to obtain a second processing result, dividing the laser point cloud data into ground point type point cloud data and non-ground point type point cloud data according to the second processing result, automatically identifying the laser point cloud data in batches by running a script according to the second processing result to obtain a second identification result, dividing the laser point cloud data into a second soil part and a second background part according to the second identification result, and retaining all second pixel points of the green vegetation part in the laser point cloud data, creating a second group canopy range boundary for each material in the cotton group in the laser point cloud data according to the second pixel points, the second soil part and the second background part, and generating a digital elevation model DEM based on the ground point type point cloud data using the second group canopy range boundary through structure from motion SfM, generating a second digital surface model DSM based on the non-ground point type point cloud data using the second group canopy range boundary through SfM, and obtaining second plant height data based on the digital elevation model DEM and the second digital surface model DSM; An acquisition module, used for acquiring the field-measured plant height data of some individual plants in the cotton field; wherein the field-measured plant height data is obtained through a measurement process from the ground to the top of the stem of the individual plant, and the time difference between the measurement time of the measurement process and the acquisition time of the RGB image data and the laser point cloud data is less than a preset time period; A training module, used to use a plurality of different machine learning algorithms to respectively train and verify the first plant height data, the second plant height data, and the field-measured plant height data to obtain final first plant height data and final second plant height data; An analysis module is used for filtering the cotton population genome variation data SNPs in the cotton field to obtain the target cotton population genome variation data SNPs whose actual data quality and accuracy are greater than before filtering, calculating the best linear unbiased estimator BLUE based on the final first plant height data and the final second plant height data, and performing GWAS analysis on the best linear unbiased estimator BLUE and the target cotton population genome variation data SNPs to obtain analysis results, and obtaining plant height gene data based on the analysis results; The first construction module is used to extract the coding sequence of the located plant height-related gene based on the plant height gene data, amplify the plant height gene using specific primers according to the coding sequence to obtain an amplified fragment, and insert the amplified fragment into a specified target vector to obtain a target gene overexpression vector, and generate gene overexpression new material data based on the target gene overexpression vector; The second construction module is used to determine the sgRNA sequence of the plant height-related gene based on the plant height gene data, insert the sgRNA sequence into the specified gene knockout vector to obtain the target gene knockout vector, and generate gene editing new material data based on the target gene knockout vector.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

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 7.

Images