High-precision phenotype analysis system for plant growth stress response

Abstract

The invention discloses a high-precision phenotype analysis system for plant growth stress response, and relates to the field of agriculture. Comprising a central analysis platform, and the central analysis platform comprises a calculation engine, a stress diagnosis model and a visual interface; the edge calculation stage comprises a data preprocessing module, a feature extraction module and a time sequence compression module; and the sensor array comprises a main control unit, a hyperspectral camera, a thermal infrared camera, a 3DDoF camera and an environment sensor group. Based on a space-time registration mechanism of transfer learning, errors are reduced, the problem of stomatal conductance modeling distortion caused by time difference is solved, a dynamic weight fusion model adjusts feature weights in real time through sensor confidence and environment fitness, illumination abrupt change and temperature and humidity fluctuation interference are eliminated, and the accuracy of stomatal conductance modeling is improved. Compared with a traditional fixed weight model, the stress identification precision is improved.

Description

Technical Field

[0001] This invention relates to the agricultural field, and more particularly to a high-precision phenotypic analysis system for plant growth stress responses. Background Technology

[0002] This high-precision phenotypic analysis system for plant growth stress responses involves the intersection of modern agricultural information technology and plant phenomics. Addressing the need for monitoring plant physiological states under stresses such as drought, salinity, and pests and diseases, the system acquires data through multispectral imaging (400-1700nm), thermal infrared imaging (8-14μm), 3D point cloud reconstruction, and a microenvironmental sensor network. It is primarily applied to crop breeding screening, precision irrigation decision-making, and large-scale field phenomics research. Multispectral technology is used for chlorophyll content analysis, thermal imaging for canopy temperature distribution, 3D reconstruction for calculating morphological parameters such as leaf area index, and environmental sensors for real-time monitoring of parameters such as photosynthetically active radiation and soil electrical conductivity.

[0003] Existing methods rely on stitching data from independent sensor modules, which does not solve the problem of spatiotemporal registration of multi-source heterogeneous data. Especially in rapidly changing field environments, the stomatal conductance model error exceeds 25% due to the time difference between acquisition of spectral and thermal imaging data, which severely limits the quantification accuracy of stress response.

[0004] Therefore, this invention proposes a high-precision phenotypic analysis system for plant growth stress response. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a high-precision phenotypic analysis system for plant growth stress responses.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A high-precision phenotypic analysis system for plant growth stress responses, including: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0007] Preferably, the environmental sensor group includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

[0008] Preferably, the working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for retransmission. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

[0009] Preferably, in step S2, data verification includes the following steps: A1: Obtain the sampling data of each sensor according to the sampling time node of the main control unit; A2: If the sampled data is unique, the verification is correct; otherwise, the verification fails. A3: For data that fails validation, delete all data in that node, and then use the formula... Replace the data at this node. P represents the data collected at the i-th time point prior to the current time point, where P is the replacement data.

[0010] Preferably, in step S2, feature extraction includes the following steps: B1: Morphological feature extraction. Morphological features are extracted based on 3D point cloud. First, RANSAC plane fitting is used to separate the leaves from the background. Then, the leaf area is calculated. Subsequently, the plant height and volume are calculated using the axial bounding box method. B2: Physiological feature extraction, using thermal imaging and hyperspectral imaging to extract the mean canopy temperature, stomatal conductance index and NDVI features respectively; B3: Stress response feature fusion, which involves performing principal component analysis to fuse the extracted features.

[0011] Preferably, in step B3, the top 5 features with 90%-95% variance are retained during feature extraction.

[0012] Preferably, in step S2, time series compression includes the following steps: C1: Important point detection is performed using a sliding window quadratic set algorithm; C2: Periodic pattern coding, which performs Fourier transform on the characteristic fluctuations of the 24-hour diurnal cycle and retains only the first 3 main frequency components; C3: Then perform Huffman coding.

[0013] Preferably, in step S2, the method for encrypting transmission includes the following steps: D1: Establish an encryption set M, which contains m encryption algorithms, and assign a unique identifier d to each encryption algorithm; D2: Divide the data into m data blocks, extract the data features of each data block, and then normalize the data features to the interval [Q, T]. D3: Establish a mapping set, d = f([Q, T]), where f() is the mapping function; D4: After normalizing the data features of the extracted data block, perform mapping processing through the mapping function f(), and then use the encryption algorithm that matches the identity identifier d of the mapping result as the encryption algorithm for the data block; In step D1, the encryption algorithm includes any combination of symmetric encryption algorithms, asymmetric encryption algorithms, hash algorithms, post-quantum encryption algorithms, and conic section algorithms. The conic section algorithm is a dynamic update algorithm, and its dynamic update method includes the following steps: D11: Establish a three-dimensional coordinate system, and create an initial cone within the three-dimensional coordinate system. Then, use the origin of the three-dimensional coordinate system as the center of the cone's base. D12: Establish mapping function Map the time range 0-24h to the interval 0-90; D13: Obtain the encryption time t, and then obtain the mapping result d according to the mapping function; D14: Generate a plane in three-dimensional space with an angle d with the xy plane to cut the initial cone, and obtain the contact cutting curve between the plane and the cone surface; D15: Obtain the cutting curve expression based on the coordinates of the coordinate system. The independent variable of the expression is the original data, and the dependent variable is the encrypted data.

[0014] In step D2, the data characteristics are any one of the following: number of characters, data block memory size, and the sequence number of the data block in the original data segmentation.

[0015] Preferably, in step S3, the calculation model for the stress response index is as follows: ,in , , These are the nutrient response index, photochemical reflectance index, and moisture slope index. The decision logic is as follows: (The text abruptly ends here, so the translation stops as well.) D1: An SRI greater than 0.7 is considered normal; D2: An alarm is triggered when SRI is less than 0.3; S3: When SRI is between 0.3 and 0.7, it is in tracking mode; In step D3, the main control unit increases the sampling frequency of the sensor array during tracking.

[0016] The beneficial effects of this invention are as follows: This invention uses a spatiotemporal registration mechanism based on transfer learning to reduce errors and solve the problem of distorted porosity modeling caused by time difference.

[0017] The dynamic weight fusion model of this invention adjusts feature weights in real time by adjusting sensor confidence and environmental adaptability, eliminating interference from sudden changes in illumination and fluctuations in temperature and humidity, and improving the accuracy of stress identification compared with the traditional fixed weight model. Attached Figure Description

[0018] Figure 1 This is a flowchart of the high-precision phenotypic analysis system for plant growth stress response proposed in this invention. Detailed Implementation

[0019] The technical solution of the present invention will be further described in detail below with reference to specific embodiments.

[0020] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "setting" should be interpreted broadly. For example, they can refer to a fixed connection or setting, a detachable connection or setting, or an integral connection or setting. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Example 1: A high-precision phenotypic analysis system for plant growth stress responses, comprising: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0022] The environmental sensor group includes a PAR sensor, a temperature and humidity sensor, a carbon dioxide sensor, and a soil EC sensor.

[0023] Example 2: A high-precision phenotypic analysis system for plant growth stress responses, comprising: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0024] The environmental sensor suite includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

[0025] The working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for retransmission. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

[0026] Example 3: A high-precision phenotypic analysis system for plant growth stress responses, including: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0027] The environmental sensor suite includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

[0028] The working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for correction. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

[0029] In step S2, data verification includes the following steps: A1: Obtain the sampling data of each sensor according to the sampling time node of the main control unit; A2: If the sampled data is unique, the verification is correct; otherwise, the verification fails. A3: For data that fails validation, delete all data in that node, and then use the formula... Replace the data at this node. P represents the data collected at the i-th time point prior to the current time point, where P is the replacement data.

[0030] In step S2, feature extraction includes the following steps: B1: Morphological feature extraction. Morphological features are extracted based on 3D point cloud. First, RANSAC plane fitting is used to separate the leaves from the background. Then, the leaf area is calculated. Subsequently, the plant height and volume are calculated using the axial bounding box method. B2: Physiological feature extraction, using thermal imaging and hyperspectral imaging to extract the mean canopy temperature, stomatal conductance index and NDVI features respectively; B3: Stress response feature fusion, which involves performing principal component analysis to fuse the extracted features.

[0031] In step B3, the top 5 features with 90%-95% variance are retained during feature extraction.

[0032] In step S2, time series compression includes the following steps: C1: Important point detection is performed using a sliding window quadratic set algorithm; C2: Periodic pattern coding, which performs Fourier transform on the characteristic fluctuations of the 24-hour diurnal cycle and retains only the first 3 main frequency components; C3: Then perform Huffman coding.

[0033] Example 4: A high-precision phenotypic analysis system for plant growth stress responses, including: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0034] The environmental sensor suite includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

[0035] The working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for correction. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

[0036] In step S2, data verification includes the following steps: A1: Obtain the sampling data of each sensor according to the sampling time node of the main control unit; A2: If the sampled data is unique, the verification is correct; otherwise, the verification fails. A3: For data that fails validation, delete all data in that node, and then use the formula... Replace the data at this node. P represents the data collected at the i-th time point prior to the current time point, where P is the replacement data.

[0037] In step S2, feature extraction includes the following steps: B1: Morphological feature extraction. Morphological features are extracted based on 3D point cloud. First, RANSAC plane fitting is used to separate the leaves from the background. Then, the leaf area is calculated. Subsequently, the plant height and volume are calculated using the axial bounding box method. B2: Physiological feature extraction, using thermal imaging and hyperspectral imaging to extract the mean canopy temperature, stomatal conductance index and NDVI features respectively; B3: Stress response feature fusion, which involves performing principal component analysis to fuse the extracted features.

[0038] In step B3, the top 5 features with 90%-95% variance are retained during feature extraction.

[0039] In step S2, time series compression includes the following steps: C1: Important point detection is performed using a sliding window quadratic set algorithm; C2: Periodic pattern coding, which performs Fourier transform on the characteristic fluctuations of the 24-hour diurnal cycle and retains only the first 3 main frequency components; C3: Then perform Huffman coding.

[0040] In step S2, the method for encrypting transmission includes the following steps: D1: Establish an encryption set M, which contains m encryption algorithms, and assign a unique identifier d to each encryption algorithm; D2: Divide the data into m data blocks, extract the data features of each data block, and then normalize the data features to the interval [Q, T]. D3: Establish a mapping set, d = f([Q, T]), where f() is the mapping function; D4: After normalizing the data features of the extracted data block, perform mapping processing through the mapping function f(), and then use the encryption algorithm that matches the identity identifier d of the mapping result as the encryption algorithm for the data block.

[0041] In step D2, the data characteristics are any one of the following: number of characters, data block memory size, and the sequence number of the data block in the original data segmentation.

[0042] In step S3, the calculation model for the stress response index is as follows: ,in , , These are the nutrient response index, photochemical reflectance index, and moisture slope index. Each has a dynamic weight.

[0043] In step S3, the decision logic is as follows: E1: An SRI greater than 0.7 is considered normal. E2: An alarm is triggered when SRI is less than 0.3; E3: When SRI is between 0.3 and 0.7, it is in tracking mode.

[0044] In step E3, the main control unit increases the sampling frequency of the sensor array during tracking.

[0045] Example 5: A high-precision phenotypic analysis system for plant growth stress responses, including: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

[0046] The environmental sensor suite includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

[0047] The working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for retransmission. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

[0048] In step S2, data verification includes the following steps: A1: Obtain the sampling data of each sensor according to the sampling time node of the main control unit; A2: If the sampled data is unique, the verification is correct; otherwise, the verification fails. A3: For data that fails validation, delete all data in that node, and then use the formula... Replace the data at this node. P represents the data collected at the i-th time point prior to the current time point, where P is the replacement data.

[0049] In step S2, feature extraction includes the following steps: B1: Morphological feature extraction. Morphological features are extracted based on 3D point cloud. First, RANSAC plane fitting is used to separate the leaves from the background. Then, the leaf area is calculated. Subsequently, the plant height and volume are calculated using the axial bounding box method. B2: Physiological feature extraction, using thermal imaging and hyperspectral imaging to extract the mean canopy temperature, stomatal conductance index and NDVI features respectively; B3: Stress response feature fusion, which involves performing principal component analysis to fuse the extracted features.

[0050] In step B3, the top 5 features with 90%-95% variance are retained during feature extraction.

[0051] In step S2, time series compression includes the following steps: C1: Important point detection is performed using a sliding window quadratic set algorithm; C2: Periodic pattern coding, which performs Fourier transform on the characteristic fluctuations of the 24-hour diurnal cycle and retains only the first 3 main frequency components; C3: Then perform Huffman coding.

[0052] In step S2, the method for encrypting transmission includes the following steps: D1: Establish an encryption set M, which contains m encryption algorithms, and assign a unique identifier d to each encryption algorithm; D2: Divide the data into m data blocks, extract the data features of each data block, and then normalize the data features to the interval [Q, T]. D3: Establish a mapping set, d = f([Q, T]), where f() is the mapping function; D4: After normalizing the data features of the extracted data block, perform mapping processing through the mapping function f(), and then use the encryption algorithm that matches the identity identifier d of the mapping result as the encryption algorithm for the data block; In step D1, the encryption algorithm includes any combination of symmetric encryption algorithms, asymmetric encryption algorithms, hash algorithms, post-quantum encryption algorithms, and conic section algorithms. The conic section algorithm is a dynamic update algorithm, and its dynamic update method includes the following steps: D11: Establish a three-dimensional coordinate system, and create an initial cone within the three-dimensional coordinate system. Then, use the origin of the three-dimensional coordinate system as the center of the cone's base. D12: Establish mapping function Map the time range 0-24h to the interval 0-90; D13: Obtain the encryption time t, and then obtain the mapping result d according to the mapping function; D14: Generate a plane in three-dimensional space with an angle d with the xy plane to cut the initial cone, and obtain the contact cutting curve between the plane and the cone surface; D15: Obtain the cutting curve expression based on the coordinates of the coordinate system. The independent variable of the expression is the original data, and the dependent variable is the encrypted data.

[0053] In step D2, the data characteristics are any one of the following: number of characters, data block memory size, and the sequence number of the data block in the original data segmentation.

[0054] In step S3, the calculation model for the stress response index is as follows: ,in , , These are the nutrient response index, photochemical reflectance index, and moisture slope index. The decision logic is as follows: (The text abruptly ends here, so the translation stops as well.) D1: An SRI greater than 0.7 is considered normal; D2: An alarm is triggered when SRI is less than 0.3; S3: When SRI is between 0.3 and 0.7, it is in tracking mode; In step D3, the main control unit increases the sampling frequency of the sensor array during tracking.

[0055] This invention uses a spatiotemporal registration mechanism based on transfer learning to reduce errors and solve the problem of distorted porosity modeling caused by time difference.

[0056] This invention proposes a dynamic weighted fusion model that adjusts feature weights in real time based on sensor confidence and environmental adaptability, eliminating interference from sudden changes in illumination and fluctuations in temperature and humidity, thus improving the accuracy of stress identification compared to traditional fixed-weight models.

[0057] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A high-precision phenotypic analysis system for plant growth stress response, characterized in that, include: A central analysis platform, which includes a computing engine, a stress diagnosis model, and a visualization interface; An edge computing node, wherein the edge computing stage includes a data preprocessing module, a feature extraction module, and a time series compression module; The sensor array includes a main control unit, a hyperspectral camera, a thermal infrared camera, a 3D ToF camera, and an environmental sensor group.

2. The high-precision phenotypic analysis system for plant growth stress response according to claim 1, characterized in that, The environmental sensor suite includes any combination of PAR sensors, temperature and humidity sensors, carbon dioxide sensors, and soil EC sensors.

3. The high-precision phenotypic analysis system for plant growth stress response according to claim 1 or 2, characterized in that, The working logic of the high-precision phenotyping system includes the following steps: S1: Data acquisition. The main control unit triggers the sensor array to synchronously acquire signals through GPIO. The data acquired by the sensor array is timestamped and then transmitted to the edge computing node via the CAN bus. S2: Data processing edge computing nodes perform verification preprocessing on the data. Data that fails verification is returned to the sensor array for retransmission. Data that passes verification is sequentially subjected to radiometric correction, feature extraction, time series compression, and then encrypted before being transmitted to the central analysis platform. S3: The central analysis platform calculates the stress response index through a computational model, and then the stress diagnosis model triggers a decision based on the calculated stress response index.

4. The high-precision phenotypic analysis system for plant growth stress response according to claim 3, characterized in that, In step S2, data verification includes the following steps: A1: Obtain the sampling data of each sensor according to the sampling time node of the main control unit; A2: If the sampled data is unique, the verification is correct; otherwise, the verification fails. A3: For data that fails validation, delete all data in that node, and then use the formula... Replace the data at this node. P represents the data collected at the i-th time point prior to the current time point, where P is the replacement data.

5. The high-precision phenotypic analysis system for plant growth stress response according to claim 3, characterized in that, In step S2, feature extraction includes the following steps: B1: Morphological feature extraction. Morphological features are extracted based on 3D point cloud. First, RANSAC plane fitting is used to separate the leaves from the background. Then, the leaf area is calculated. Subsequently, the plant height and volume are calculated using the axial bounding box method. B2: Physiological feature extraction, using thermal imaging and hyperspectral imaging to extract the mean canopy temperature, stomatal conductance index and NDVI features respectively; B3: Stress response feature fusion, which involves performing principal component analysis to fuse the extracted features.

6. The high-precision phenotypic analysis system for plant growth stress response according to claim 5, characterized in that, In step B3, the top 5 features with 90%-95% variance are retained during feature extraction.

7. The high-precision phenotypic analysis system for plant growth stress response according to claim 3, characterized in that, In step S2, time series compression includes the following steps: C1: Important point detection is performed using a sliding window quadratic set algorithm; C2: Periodic pattern coding, which performs Fourier transform on the characteristic fluctuations of the 24-hour diurnal cycle and retains only the first 3 main frequency components; C3: Then perform Huffman coding.

8. The high-precision phenotypic analysis system for plant growth stress response according to claim 3, characterized in that, In step S2, the method for encrypting transmission includes the following steps: D1: Establish an encryption set M, which contains m encryption algorithms, and assign a unique identifier d to each encryption algorithm; D2: Divide the data into m data blocks, extract the data features of each data block, and then normalize the data features to the interval [Q, T]. D3: Establish a mapping set, d = f([Q, T]), where f() is the mapping function; D4: After normalizing the data features of the extracted data block, perform mapping processing through the mapping function f(), and then use the encryption algorithm that matches the identity identifier d of the mapping result as the encryption algorithm for the data block; In step D1, the encryption algorithm includes any combination of symmetric encryption algorithms, asymmetric encryption algorithms, hash algorithms, post-quantum encryption algorithms, and conic section algorithms. The conic section algorithm is a dynamic update algorithm, and its dynamic update method includes the following steps: D11: Establish a three-dimensional coordinate system, and create an initial cone within the three-dimensional coordinate system. Then, use the origin of the three-dimensional coordinate system as the center of the cone's base. D12: Establish mapping function Map the time range 0-24h to the interval 0-90; D13: Obtain the encryption time t, and then obtain the mapping result d according to the mapping function; D14: Generate a plane in three-dimensional space with an angle d with the xy plane to cut the initial cone, and obtain the contact cutting curve between the plane and the cone surface; D15: Obtain the cutting curve expression based on the coordinates of the coordinate system. The independent variable of the expression is the original data, and the dependent variable is the encrypted data.

9. The high-precision phenotypic analysis system for plant growth stress response according to claim 8, characterized in that, In step D2, the data characteristics are any one of the following: number of characters, data block memory size, and the sequence number of the data block in the original data segmentation.

10. The high-precision phenotypic analysis system for plant growth stress response according to claim 3, characterized in that, In step S3, the calculation model for the stress response index is as follows: ,in , , These are the nutrient response index, photochemical reflectance index, and moisture slope index. The decision logic is as follows: (The text abruptly ends here, so the translation stops as well.) D1: An SRI greater than 0.7 is considered normal; D2: An alarm is triggered when SRI is less than 0.3; S3: When SRI is between 0.3 and 0.7, it is in tracking mode; In step D3, the main control unit increases the sampling frequency of the sensor array during tracking.

Images