Cotton phenotype dynamic monitoring method and device, medium and equipment

Abstract

The invention discloses a cotton phenotype dynamic monitoring method and device, a medium and equipment, and relates to the field of crop growth monitoring, and the method comprises the steps: carrying out the correlation analysis of cotton phenotype characteristics, and determining the representation information of the plant height daily increment and the top five-node length as the shrinkage joint safety sensitive phenotype characteristics; based on the obtained cotton multispectral image in the test area, extracting the daily increment of the plant height, and constructing an inversion model of the daily increment of the plant height by performing unary linear regression on the daily increment of the plant height; constructing a machine learning model; the method comprises the following steps: measuring the total length of five internodes below the top of a cotton stem in a test area, determining representation information of the length of the five internodes at the top, and training a machine learning model by combining a cotton multispectral image to obtain a prediction model of the length of the five internodes at the top; a cotton multispectral image is acquired in real time, and according to a top five-node length prediction model and a plant height daily increment inversion model, a contraction joint safety sensitive phenotype feature prediction result is determined, so that contraction joint safety sensitive phenotype monitoring is realized.

Description

Technical Field

[0001] The invention relates to the technical field of crop growth monitoring, and in particular to a method, device, medium and equipment for dynamically monitoring cotton phenotypes. Background Art

[0002] With the rapid development of cotton cultivation industry, the full mechanization, simplification and precision of cotton field operations have become a trend. The quantitative relationship between sensor information in cotton fields and the growth and development of cotton plants and the input of pesticides is the basis of precision operations. Some studies have used vehicle-mounted sensors to obtain phenotypes such as biomass, plant height, and SPAD value of cotton fields, and combined with the NDVI (Normalized Difference Vegetation Index) distribution to attempt to apply MC variables at a regional scale (Foote et al., 2016; Motomiya et al., 2014; Trevisan et al., 2018; Vellidis et al., 2009). However, there are limitations in using a single vegetation index (VI) such as NDVI to guide MC application. Spectral saturation occurs in the late stage of cotton fields (Pabuayon et al., 2019), which is easily affected by environmental factors and difficult to accurately evaluate the effect of MC application.

[0003] During the growth of cotton, Mepiquat Chloride (MC) is widely used in cotton planting as a growth regulator (Feng et al., 2024). MC can monitor cotton growth and development, control cotton plant height and population structure by adjusting cotton hormone levels (Wang et al., 2014), and coordinate vegetative growth and reproductive growth.

[0004] In recent years, the combination of UAV technology and multispectral imaging systems has opened up new ways for the inversion of crop phenotypic characteristics, expanded the spatial scope and temporal frequency of data collection, and significantly improved the accuracy and precision of crop growth status assessment by comprehensively analyzing multi-dimensional information (Maes and Steppe, 2019). UAVs equipped with multispectral cameras can capture subtle differences in the reflected light of crop leaves in different bands. Spectral features include specific band values ​​in multispectral images and constructed spectral indices. These spectral features are directly related to the leaf area index (LAI), chlorophyll content, water status, nutrient levels, and early signs of pests and diseases of crops (Chen et al., 2021; Liu et al., 2023a; Risal et al., 2024; B. Zhang et al., 2024; C. Zhang et al., 2024), providing key indicators for understanding crop growth vitality and stress response. In addition to spectral features, high-resolution images acquired by drones also allow the extraction of rich texture features, which have been widely used to estimate physiological parameters such as crop AGB, LAI, and crop chlorophyll content (Liu et al., 2019; Wang et al., 2022; Yuan et al., 2023), reflecting the heterogeneity of crop growth and possible environmental stress responses. At the same time, the point cloud data generated by drone photogrammetry provides three-dimensional geometric structural information of the crop canopy. By analyzing structural and geometric features such as crop height, volume, canopy density, point cloud curvature, and planarity (Dhakal et al., 2023; Meacham-Hensold et al., 2019; Tsoulias et al., 2020), the understanding of crop plant type, biomass, and population competition can be further refined, which is of great value for optimizing planting patterns and precision pesticide application.

[0005] Although the above studies involve the inversion of many crop phenotypes, no targeted research has been conducted on the MC-sensitive phenotype in cotton cultivation, which has certain limitations for MC application decisions, and thus affects the accuracy of monitoring the phenotype of cotton that is sensitive to thiamethoxam. Summary of the invention

[0006] The present invention provides a method, device, medium and equipment for dynamically monitoring cotton phenotypes, which are used to solve the above-mentioned problem existing in the prior art, that is, how to accurately dynamically monitor the phenotype of cotton sensitive to joint shrinkage in the prior art. The present invention provides a method for dynamically monitoring cotton phenotypes, which comprises:

[0007] The Spearman correlation coefficient method was used to analyze the correlation of cotton phenotypic characteristics, and the characterization information of daily increment of plant height and length of top 5 internodes were determined as the sensitive phenotypic characteristics of stunted joint anesthesia.

[0008] Acquire multispectral images of cotton in the test area;

[0009] Based on the multispectral images of cotton in the experimental area, the daily increment of plant height was extracted, and the inversion model of daily increment of plant height was constructed by performing univariate linear regression on the daily increment of plant height.

[0010] Construct a machine learning model; measure the total length of the five internodes below the top of the main stem of cotton in the test area, determine the characterization information of the length of the top five internodes, and train the machine learning model to obtain a prediction model for the length of the top five internodes by combining the multispectral images of cotton in the corresponding period;

[0011] By acquiring multispectral images of cotton in real time, the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus are determined based on the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and the phenotypic characteristics sensitive to the shortened joint annuls are monitored.

[0012] Also includes:

[0013] The multispectral images of cotton in the experimental area were subjected to radiometric calibration, image alignment, construction of dense point clouds and ortho-mosaic processing.

[0014] Optionally, extracting the daily increment of plant height based on the multispectral image of cotton in the test area specifically includes:

[0015] Based on the multi-spectral image of cotton in the test area, the image of the test area was obtained. According to the image of the test area, the daily increase of plant height was obtained using the following formula:

[0016]

[0017]

[0018] Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

[0019] Optionally, extracting the daily increment of plant height based on the multispectral image of cotton in the test area specifically includes:

[0020] Based on the multispectral image of cotton in the test area, the quantiles of the test area images were obtained, the best estimated plant height quantiles of the corresponding growth period were obtained, the difference in plant heights in adjacent periods was determined, and the daily increment of plant height was obtained using the following formula:

[0021]

[0022]

[0023] Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

[0024] The present invention provides a cotton phenotype dynamic monitoring device, comprising:

[0025] An extraction module is used to perform correlation analysis on cotton phenotypic characteristics using the Spearman correlation coefficient method, and determine the characterization information of plant height daily increment and top 5 internode length as the phenotypic characteristics of the sensitivity to joint shortening;

[0026] An acquisition module, used to acquire multispectral images of cotton in the test area;

[0027] The plant height daily increment inversion model construction module is used to extract the plant height daily increment based on the multispectral image of cotton in the test area, and to construct the plant height daily increment inversion model by performing univariate linear regression on the plant height daily increment;

[0028] The module for building a prediction model for the length of the top 5 internodes is used to build a machine learning model. The total length of the 5 internodes below the top of the main stem of cotton in the test area is measured to determine the characterization information of the length of the top 5 internodes. Combined with the multispectral images of cotton in the corresponding period, the machine learning model is trained to obtain a prediction model for the length of the top 5 internodes.

[0029] The monitoring module is used to obtain cotton multispectral images in real time, determine the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus according to the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and monitor the phenotypic characteristics sensitive to the shortened joint annuls.

[0030] The present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned cotton phenotype dynamic monitoring method is implemented.

[0031] The present invention provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for dynamically monitoring cotton phenotypes when executing the program.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method for dynamically monitoring cotton phenotypes, which selects characterization information of daily plant height increment and length of top 5 internodes as the joint-cutting sensitive phenotype by performing correlation analysis on cotton phenotypes; extracts daily plant height increment based on cotton multispectral images, constructs an inversion model of daily plant height increment, and based on the characterization information of the length of top 5 internodes, adopts a machine learning model to predict the length of top 5 internodes, which can obtain the characterization information of daily plant height increment and length of top 5 internodes, two types of prediction results of joint-cutting sensitive phenotype characteristics by real-time input of multispectral images, thereby realizing accurate monitoring of joint-cutting sensitive phenotypic characteristics in cotton phenotypes. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0034] Figure 1 A flowchart of a method for dynamically monitoring cotton phenotypes provided by an embodiment of the present invention;

[0035] Figure 2 A schematic diagram of a computer device for a method for dynamically monitoring cotton phenotypes provided in an embodiment of the present invention.

[0036] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0039] Figure 1 is a flow chart of a method for dynamically monitoring cotton phenotypes provided by an embodiment of the present invention, such as Figure 1As shown, this embodiment shows a method for dynamically monitoring cotton phenotypes, comprising:

[0040] S1: The Spearman correlation coefficient method was used to conduct correlation analysis on cotton phenotypic characteristics, and the characterization information of daily plant height increment and length of the top 5 internodes were determined as phenotypic characteristics sensitive to truncation.

[0041] Exemplarily, cotton phenotypic characteristics may include plant height (PH), top 5 internode length (TOP5), daily plant height increment (PHI), node above white flower (NAWF), leaf area index (LAI), aggregation index (ACF), canopy openness (DIFN), mean leaf inclination angle (MTA), porosity (GAPS) and SPAD value.

[0042] S2: Obtain multispectral images of cotton in the test area.

[0043] For example, 10 representative cotton plants can be selected in the test area for marking, and the cotton phenotypic data can be measured 0d, 9d, 17d, 26d, and 31d after the application of cyclohexan from the bud stage to the full flowering stage. The drone platform can be the DJI Phantom 4 Multispectral Edition to collect multispectral images of the cotton canopy.

[0044] Optionally, the multispectral image is subjected to radiometric calibration, image alignment, construction of dense point cloud, and orthomosaic processing.

[0045] S3: Based on the multispectral images of cotton in the experimental area, the daily increment of plant height was extracted, and an inversion model of the daily increment of plant height was constructed by performing univariate linear regression on the daily increment of plant height.

[0046] Exemplarily, the test area image DSM generated by the acquired drone multispectral image is used as the digital terrain model DTM and as the surface reference plane for subsequent plant height data extraction. The calibrated test area image is obtained by calculating the difference between the test area image DSM and the digital terrain model DTM at each cotton period.

[0047] Optionally, an image of the test area is obtained based on the multi-spectral image of cotton in the test area, and the daily increment of plant height is obtained using the following formula based on the image of the test area:

[0048]

[0049]

[0050] Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

[0051] In addition, the quantiles of the images in the test area can be obtained through the multispectral images of cotton in the test area, the best estimated plant height quantiles of the corresponding growth period can be obtained, the difference in plant heights in adjacent periods can be determined, and the daily increment of plant height can be obtained using the following formula:

[0052]

[0053]

[0054] Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

[0055] S4: Construct a machine learning model; by measuring the total length of the 5 internodes below the top of the cotton main stem in the test area, determine the characterization information of the length of the top 5 internodes, and combine the multispectral images of cotton in the corresponding period to train the machine learning model to obtain a prediction model for the length of the top 5 internodes.

[0056] Optionally, six machine learning algorithms including partial least squares regression, random forest regression and support vector regression, Gaussian process regression, gradient boosted tree or cross-validated ridge regression were used to obtain the top 5 internode length prediction models.

[0057] S5: By acquiring cotton multispectral images in real time, the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus are determined based on the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and the phenotypic characteristics sensitive to the shortened joint annuls are monitored.

[0058] The above is a method for dynamically monitoring cotton phenotypes provided by one or more embodiments of this specification. Based on the same idea, this specification also provides a corresponding device for dynamically monitoring cotton phenotypes, including:

[0059] An extraction module is used to perform correlation analysis on cotton phenotypic characteristics using the Spearman correlation coefficient method, and determine the characterization information of plant height daily increment and top 5 internode length as the phenotypic characteristics of the sensitivity to joint shortening;

[0060] An acquisition module, used to acquire multispectral images of cotton in the test area;

[0061] The plant height daily increment inversion model construction module is used to extract the plant height daily increment based on the multispectral image of cotton in the test area, and to construct the plant height daily increment inversion model by performing univariate linear regression on the plant height daily increment;

[0062] The module for building a prediction model for the length of the top 5 internodes is used to build a machine learning model. The total length of the 5 internodes below the top of the main stem of cotton in the test area is measured to determine the characterization information of the length of the top 5 internodes. Combined with the multispectral images of cotton in the corresponding period, the machine learning model is trained to obtain a prediction model for the length of the top 5 internodes.

[0063] The monitoring module is used to obtain cotton multispectral images in real time, determine the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus according to the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and monitor the phenotypic characteristics sensitive to the shortened joint annuls.

[0064] The specific definition of the cotton phenotype dynamic monitoring device can be found in the above-mentioned definition of the cotton phenotype dynamic monitoring method, which will not be repeated here. Each module in the above-mentioned cotton phenotype dynamic monitoring device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0065] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the cotton phenotype dynamic monitoring method provided above.

[0066] The present invention also provides Figure 2 The structural diagram of the computer device shown in FIG. Figure 2 As shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the cotton phenotype dynamic monitoring method provided in the above embodiment.

[0067] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0068] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for dynamically monitoring cotton phenotypes, characterized in that: include: The Spearman correlation coefficient method was used to analyze the correlation of cotton phenotypic characteristics, and the characterization information of daily increment of plant height and length of top 5 internodes were determined as the sensitive phenotypic characteristics of stunted joint anesthesia. Acquire multispectral images of cotton in the test area; Based on the multispectral images of cotton in the experimental area, the daily increment of plant height was extracted, and the inversion model of daily increment of plant height was constructed by performing univariate linear regression on the daily increment of plant height. Construct a machine learning model; measure the total length of the five internodes below the top of the main stem of cotton in the test area, determine the characterization information of the length of the top five internodes, and train the machine learning model to obtain a prediction model for the length of the top five internodes by combining the multispectral images of cotton in the corresponding period; By acquiring multispectral images of cotton in real time, the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus are determined based on the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and the phenotypic characteristics sensitive to the shortened joint annuls are monitored.

2. The method for dynamically monitoring cotton phenotypes according to claim 1, characterized in that: The method further comprises: The multispectral images of cotton in the experimental area were subjected to radiometric calibration, image alignment, construction of dense point clouds and ortho-mosaic processing.

3. The method for dynamically monitoring cotton phenotypes according to claim 1, characterized in that: The method of extracting the daily increment of plant height based on the multispectral image of cotton in the test area specifically includes: Based on the multi-spectral image of cotton in the test area, the image of the test area was obtained. According to the image of the test area, the daily increase of plant height was obtained using the following formula: Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

4. The method for dynamically monitoring cotton phenotypes according to claim 1, characterized in that: The method of extracting the daily increment of plant height based on the multispectral image of cotton in the test area specifically includes: Based on the multispectral image of cotton in the test area, the quantiles of the test area images were obtained, the best estimated plant height quantiles of the corresponding growth period were obtained, the difference in plant heights in adjacent periods was determined, and the daily increment of plant height was obtained using the following formula: Among them, n is the number of image data points in the test area, p is the percentile, k is the index number after the data set is sorted, DSM1 is the image of the early test area, DSM2 is the image of the later test area, DAYS is the number of days between the two images, and PHI is the daily increase in plant height.

5. A cotton phenotype dynamic monitoring device, characterized in that: include: An acquisition module, used to acquire multispectral images of cotton in the test area; An extraction module is used to perform correlation analysis on cotton phenotypic characteristics using the Spearman correlation coefficient method, and determine the characterization information of plant height daily increment and top 5 internode length as the phenotypic characteristics of the sensitivity to joint shortening; The plant height daily increment inversion model construction module is used to extract the plant height daily increment based on the multispectral image of cotton in the test area, and to construct the plant height daily increment inversion model by performing univariate linear regression on the plant height daily increment; A module for constructing a prediction model for the length of the top five internodes, which is used to determine the characterization information of the length of the top five internodes by obtaining the total length of the five internodes below the top of the cotton main stem, and to construct a prediction model for the length of the top five internodes by using a machine learning model; The monitoring module is used to obtain cotton multispectral images in real time, determine the prediction results of the phenotypic characteristics that are sensitive to the shortened joint annulus through the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and monitor the phenotypic characteristics that are sensitive to the shortened joint annuls. An extraction module is used to perform correlation analysis on cotton phenotypic characteristics using the Spearman correlation coefficient method, and determine the characterization information of plant height daily increment and top 5 internode length as the phenotypic characteristics of the sensitivity to joint shortening; An acquisition module, used to acquire multispectral images of cotton in the test area; The plant height daily increment inversion model construction module is used to extract the plant height daily increment based on the multispectral image of cotton in the test area, and to construct the plant height daily increment inversion model by performing univariate linear regression on the plant height daily increment; The module for building a prediction model for the length of the top 5 internodes is used to build a machine learning model. The total length of the 5 internodes below the top of the main stem of cotton in the test area is measured to determine the characterization information of the length of the top 5 internodes. Combined with the multispectral images of cotton in the corresponding period, the machine learning model is trained to obtain a prediction model for the length of the top 5 internodes. The monitoring module is used to obtain cotton multispectral images in real time, determine the prediction results of the phenotypic characteristics sensitive to the shortened joint annulus according to the prediction model of the top 5 internode length and the inversion model of the daily increase in plant height, and monitor the phenotypic characteristics sensitive to the shortened joint annuls.

6. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for dynamically monitoring cotton phenotypes according to any one of claims 1 to 4 is implemented.

7. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for dynamically monitoring cotton phenotypes as claimed in any one of claims 1 to 4 is implemented.

Images