Device and method for identifying drought resistance of corn

Abstract

The invention discloses a corn drought resistance identification device and method, and belongs to the technical field of agricultural scientific research. According to the method, the problems of inaccurate environment control, long identification period, poor repeatability and the like of an existing method are solved, the drought resistance index of the corn is calculated through the random forest model, the drought resistance grade of the corn is divided, and the drought resistance index threshold value is set; the drought resistance indexes of the corn in different growth periods are analyzed, the drought resistance index threshold value is dynamically adjusted according to the corn growth data monitored in real time, and the drought resistance of the corn in different growth periods can be reflected more accurately; feature importance scores are calculated through a random forest model, feature variables are sorted, and it can be determined that certain features make the greatest contribution to the drought resistance index; based on feature importance analysis, suggestions on corn drought resistance research and breeding are provided, researchers can be helped to optimize breeding strategies, errors caused by subjective judgment are reduced, and the breeding efficiency of drought-resistant varieties is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural science research, and in particular to a device and method for identifying drought resistance of corn. Background Art

[0002] Corn is one of the world's important food crops, and its growth and yield are greatly affected by environmental stresses such as drought. With the intensification of climate change and water shortage problems, the development and screening of drought-resistant corn varieties has become an important direction of agricultural science research.

[0003] Traditional methods for identifying corn drought resistance mostly rely on natural drought conditions in the field or artificial drought treatment, but these methods have problems such as inaccurate environmental control, long identification cycle, and poor repeatability.

[0004] Therefore, the existing demand is not met, and we have proposed a device and method for identifying drought resistance of corn. Summary of the Invention

[0005] The present invention aims to provide a device and method for identifying drought resistance of corn. The device and method calculate the drought resistance index of corn through a random forest model, classify the drought resistance grades of corn and set drought resistance index thresholds. The device analyzes the drought resistance index of corn in different growth stages and dynamically adjusts the drought resistance index thresholds according to real-time monitored corn growth data, thereby more accurately reflecting the drought resistance of corn in different growth stages. The device calculates feature importance scores through a random forest model and sorts feature variables, thereby clarifying that certain features have the greatest contribution to the drought resistance index. Based on the feature importance analysis, suggestions for corn drought resistance research and breeding are proposed, thereby helping researchers optimize breeding strategies, reduce errors caused by subjective judgment, and improve the breeding efficiency of drought-resistant varieties, thereby solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A corn drought resistance identification device, comprising: a plant phenotype monitoring unit, a soil moisture monitoring unit and a drought resistance identification unit;

[0008] The plant phenotype monitoring unit is configured to process and analyze corn plant images based on multispectral imaging technology to obtain phenotypic parameters of the corn plants, the phenotypic parameters including leaf area, plant height, and chlorophyll content;

[0009] The soil moisture monitoring unit is configured to install soil moisture sensors at different depths in the corn planting area, automatically collect soil moisture data at regular intervals, and obtain soil moisture parameters; based on temperature sensors, humidity sensors, and light sensors, it monitors air temperature, humidity, and light intensity in real time;

[0010] The drought resistance identification unit is configured to integrate phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, calculate the drought resistance index of corn based on a random forest model, and obtain a corn drought resistance identification result.

[0011] Furthermore, the drought resistance identification unit includes:

[0012] A grading module is configured to extract the drought resistance index of all corn varieties or treatments from the prediction results of the random forest model, standardize the drought resistance index, classify the corn drought resistance grades using the K-means clustering algorithm, and set a drought resistance index threshold for each grade;

[0013] The dynamic adjustment module is configured to dynamically adjust the drought resistance index threshold according to the differences in the growth period of corn.

[0014] Furthermore, the dynamic adjustment module includes:

[0015] The strategy formulation module is configured to analyze the distribution of drought resistance index of corn at different growth stages, including minimum, maximum, average, and standard deviation indicators. Based on the physiological characteristics and drought resistance performance of corn at different growth stages, the initial drought resistance index threshold is set for each growth stage, and the drought resistance index threshold is dynamically adjusted based on real-time monitored corn growth data.

[0016] The result verification module is configured to verify the effect of the dynamic adjustment threshold by comparing the drought resistance grade classification results before and after the adjustment; and further optimize the dynamic adjustment strategy based on the verification results.

[0017] Furthermore, the drought resistance identification unit further includes:

[0018] A feature analysis module is configured to sort the phenotypic parameters of corn plants, soil moisture, temperature and humidity data, and light intensity from high to low according to the feature importance score, and obtain the feature variable that contributes most to the prediction of drought resistance index;

[0019] The feature ranking module is configured to list the importance scores of all feature variables based on the prediction results of random forests, focusing on analyzing the top-ranked features and explaining their contribution to the drought resistance index; based on the feature importance analysis, it puts forward suggestions for maize drought resistance research and breeding.

[0020] Furthermore, the drought resistance identification unit further includes:

[0021] a feature extraction module configured to collect historical data as training data, including samples with known drought resistance indexes, and extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity from the samples;

[0022] The data processing module is configured to perform standardization on all characteristic variables to eliminate the effects of different dimensions and orders of magnitude.

[0023] Furthermore, the drought resistance identification unit further includes:

[0024] The model building module is configured to randomly select samples from the training data and randomly select a subset of features each time a node is split. A decision tree is constructed for each sample set until the maximum depth or minimum number of samples is reached. The prediction results of all decision trees are averaged to obtain the final drought resistance index.

[0025] The model validation module is configured to validate the random forest model using samples that did not participate in the training of each tree and measure the error between the predicted value and the actual value.

[0026] Furthermore, the plant phenotype monitoring unit includes:

[0027] an image acquisition module configured to use a high-resolution camera to capture images of corn plants from multiple angles, including front, side, and top views;

[0028] The parameter acquisition module is configured to use an image segmentation algorithm to separate leaves and background for calculating leaf area, measure plant height through a feature point detection algorithm, and calculate chlorophyll content using spectral analysis technology.

[0029] Furthermore, the soil moisture monitoring unit includes:

[0030] a parameter analysis module configured to obtain soil moisture data, temperature and humidity data, and light intensity at different depths, calculate average soil moisture content, daily change rate of soil moisture content, and soil moisture gradient, and calculate the cumulative amount of daily average air temperature, average air humidity, and light intensity;

[0031] The parameter fusion module is configured to fuse soil moisture data, temperature and humidity data, and light intensity to form comprehensive data.

[0032] Furthermore, the device further comprises: a visualization terminal;

[0033] The visualization terminal is configured to display charts of soil moisture, environmental parameters, and phenotypic parameters in real time at each stage through the user interface, generate data comparison charts between different stages, and intuitively display the differences between different stages; and based on the experimental results, generate visualization charts for drought resistance evaluation to intuitively display the differences in drought resistance of different varieties.

[0034] A method for identifying drought resistance of corn comprises the following steps:

[0035] Collect corn plant images from multiple angles and calculate leaf area, plant height, and chlorophyll content to obtain phenotypic parameters;

[0036] Automatically collect soil moisture data at different depths in corn planting areas at regular intervals, and monitor air temperature, humidity, and light intensity in real time;

[0037] Extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, and rank the characteristic variables from high to low according to the characteristic importance score;

[0038] Refer to historical data to classify corn drought resistance levels, and set drought resistance index thresholds for each level based on the different growth stages of corn;

[0039] The random forest model is used to calculate the current drought resistance index of corn. Combined with the current growth period of corn, it is determined whether the drought resistance index is within the drought resistance threshold range and whether the current drought resistance index of corn needs to be adjusted.

[0040] Combined with the current ranking results of corn characteristics importance, field management suggestions are provided for current corn planting.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. In the present invention, the drought resistance index of corn is calculated by a random forest model, the K-means clustering algorithm is used to divide the corn drought resistance level, and a drought resistance index threshold is set for each level; by analyzing the distribution of the drought resistance index of corn in different growth stages, an initial drought resistance index threshold is set for each growth stage; and the drought resistance index threshold is dynamically adjusted according to real-time monitored corn growth data, which can more accurately reflect the drought resistance of corn at different growth stages, so as to more reasonably formulate a drought resistance identification strategy.

[0043] 2. In the present invention, the feature importance score is calculated by the random forest model, and the feature variables are sorted to clarify which features contribute most to the drought resistance index; based on the feature importance analysis, suggestions for maize drought resistance research and breeding are put forward, which can help researchers optimize breeding strategies, reduce errors caused by subjective judgment, and improve the efficiency of breeding drought-resistant varieties. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a diagram showing the composition of a maize drought resistance identification module according to the present invention;

[0045] Figure 2 The figure is a flow chart of the method for identifying drought resistance of corn of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0047] In order to solve the technical problems of the existing technology, traditional corn drought resistance identification methods mostly rely on natural drought conditions in the field or artificial drought treatment, but these methods have the problems of imprecise environmental control, long identification cycle, poor repeatability, etc. Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0048] A device for identifying drought resistance of corn comprises a plant phenotype monitoring unit, a soil moisture monitoring unit, a drought resistance identification unit and a visualization terminal.

[0049] The plant phenotypic monitoring unit is configured to process and analyze corn plant images based on multispectral imaging technology to obtain phenotypic parameters of the corn plants, including leaf area, plant height, and chlorophyll content. The plant phenotypic monitoring unit includes:

[0050] The image acquisition module is configured to use a high-resolution camera to collect images of corn plants from multiple angles, including a front view for observing the overall morphology of the plant and leaf distribution, a side view for measuring plant height and observing the longitudinal growth of the plant, and a top view for calculating leaf area and observing the planar distribution of leaves. In specific implementation, the camera is installed on an adjustable bracket to adapt to corn plants of different heights. The time frequency is set at 09:00-11:00 every day to avoid strong sunlight at noon.

[0051] The parameter acquisition module is configured to use an image segmentation algorithm to separate leaves and background, and combined with camera calibration parameters, convert pixel area into actual area; measure plant height through a feature point detection algorithm: base: locate the intersection of the soil surface and the stem through the HSV color space; top: obtain the highest valid pixel point through morphological top hat transformation, combine the base and the top to obtain plant height; and use spectral analysis technology to calculate chlorophyll content, including calculating the TCARI and OSAVI indices, and estimate the chlorophyll content based on the spectral index and calibration coefficient.

[0052] The soil moisture monitoring unit is configured to install soil moisture sensors at different depths in the corn planting area, such as 10cm, 20cm, and 30cm. It automatically collects soil moisture data and obtains soil moisture parameters at regular intervals. The soil moisture sensor uses time domain reflectometry technology to accurately measure soil moisture content. Based on temperature sensors, humidity sensors, and light sensors, it monitors air temperature, humidity, and light intensity in real time to ensure data representativeness. The soil moisture monitoring unit includes:

[0053] The parameter analysis module is configured to set the soil moisture sensor to collect data every 15 minutes, and the temperature and humidity sensor and light sensor to collect data once an hour, so as to obtain soil moisture data, temperature and humidity data and light intensity at different depths, calculate the average soil moisture content, the daily change rate of soil moisture content reflecting the dynamic changes of soil moisture, and the soil moisture gradient reflecting the vertical distribution of moisture in the soil, and calculate the daily average air temperature, average air humidity and cumulative light intensity.

[0054] The parameter fusion module is configured to fuse soil moisture data, temperature and humidity data, and light intensity to form comprehensive data, and store them in a structured data format for subsequent drought resistance assessment.

[0055] The drought resistance identification unit is configured to integrate phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, calculate the drought resistance index of corn based on a random forest model, and obtain the drought resistance identification result of corn. The drought resistance identification unit includes:

[0056] The feature extraction module is configured to collect historical data, organize the historical data into a unified format to ensure the consistency and integrity of the data, and use it as training data, including samples with known drought resistance indexes, to extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity in the samples; store the extracted characteristic variables in a structured data format to facilitate subsequent processing and analysis; and regularly back up the extracted characteristic variables to prevent data loss.

[0057] The data processing module is configured to standardize all feature variables, identify and process outliers in the data, ensure data accuracy, fill missing values ​​in the data, and ensure data integrity; standardize or normalize each feature variable to eliminate the influence of different dimensions and orders of magnitude, and ensure data consistency and comparability.

[0058] The feature analysis module is configured to sort the phenotypic parameters, soil moisture, temperature and humidity data, and light intensity of the corn plants from high to low according to the feature importance score, and obtain the feature variables that contribute most to the prediction of the drought resistance index; for example: a larger leaf area may indicate that the plant can still maintain a good photosynthesis capacity under drought conditions; plant height may reflect the growth vitality of the plant, and a higher plant height may indicate that the plant can still grow normally under drought conditions; a higher chlorophyll content may indicate that the plant has a higher photosynthesis efficiency and a stronger tolerance to drought stress; the dynamic changes in soil moisture directly affect the water supply of the plant, and a lower soil moisture content may indicate that the plant is in a drought stress state; a higher temperature may aggravate drought stress and affect the transpiration and water use efficiency of the plant; a lower air humidity may increase the water loss of the plant; and the light intensity may affect the photosynthesis and water use efficiency of the plant.

[0059] The feature ranking module, configured based on the prediction results of a random forest, lists the importance scores of all feature variables, focusing on analyzing the top-ranked features and explaining their contribution to the drought resistance index. Based on the feature importance analysis, it provides recommendations for corn drought resistance research and breeding. For example: Parent selection: Select varieties with higher drought resistance indexes as parents for hybrid breeding; Gene editing: Introduce known drought resistance genes through gene editing for varieties with lower drought resistance; Marker-assisted selection: Use molecular marker technology to screen gene loci related to drought resistance for marker-assisted selection breeding. Field management recommendations: Based on the current corn feature importance ranking results, field management recommendations are provided for current corn planting. For example: Irrigation management: Optimize irrigation strategies based on the dynamic changes in soil moisture and environmental parameters. Planting density: Adjust planting density based on the variety's drought resistance and growth characteristics.

[0060] The beneficial effects achieved by the above content are: by calculating the feature importance score through the random forest model and sorting the feature variables, it can be determined which features contribute most to the drought resistance index; based on the feature importance analysis, suggestions for maize drought resistance research and breeding are made, which can help researchers optimize breeding strategies, reduce errors caused by subjective judgment, and improve the efficiency of breeding drought-resistant varieties.

[0061] The model building module is configured to randomly select samples from the training data to ensure that the training data of each decision tree is diverse; each time a node is split, a part of the features is randomly selected to reduce the correlation between features and improve the robustness of the model; a decision tree is constructed for each sample set until the maximum depth or minimum number of samples is reached to prevent overfitting and ensure the reasonable complexity of the decision tree; the prediction results of all decision trees are averaged to obtain the final drought resistance index; the trained random forest model is stored locally for subsequent verification and application.

[0062] The model validation module is configured to validate the random forest model using samples that did not participate in the training of each tree, standardize the validation data to ensure consistency with the training data; use the trained random forest model to predict the validation data, calculate the mean square error, determination coefficient and mean absolute error between the predicted value and the actual value, measure the error between the predicted value and the actual value based on the error index, and evaluate the performance of the model.

[0063] The classification module is configured to extract the drought resistance index of all corn varieties or treatments from the prediction results of the random forest model, standardize the drought resistance index, and use the K-means clustering algorithm to classify the corn drought resistance level. The appropriate number of clusters K is selected according to actual needs, for example, K = 3 (low drought resistance, medium drought resistance, high drought resistance), and the center of each cluster is calculated as the threshold of the drought resistance level, thereby setting the drought resistance index threshold for each level, as shown in the following table:

[0064]

[0065]

[0066] Table 1. Corn drought resistance grade table

[0067] The dynamic adjustment module is configured to dynamically adjust the drought resistance index threshold according to the differences in the growth period of corn. The dynamic adjustment module includes:

[0068] The strategy formulation module is configured to analyze the distribution of drought resistance index of corn in different growth stages, including minimum, maximum, average, and standard deviation indicators; set the initial drought resistance index threshold for each growth stage based on the physiological characteristics and drought resistance performance of corn in different growth stages, and dynamically adjust the drought resistance index threshold based on real-time monitored corn growth data; for example:

[0069] Seedling stage: Corn is more sensitive to drought stress during the seedling stage, and the drought resistance index threshold is set lower to more strictly screen drought-resistant varieties; for example: low drought resistance DTI < 0.5, medium drought resistance 0.5 ≤ DTI < 1.0, and high drought resistance DTI ≥ 1.0.

[0070] Jointing stage: Corn grows vigorously during the jointing stage, its tolerance to drought is enhanced, and the drought resistance index threshold is appropriately increased; for example: low drought resistance DTI < 0.6, medium drought resistance 0.6 ≤ DTI < 1.2, high drought resistance DTI ≥ 1.2.

[0071] Heading stage: The heading stage is a critical period for corn growth and is very sensitive to drought stress. The drought resistance index threshold is lowered again; for example: low drought resistance DTI < 0.4, medium drought resistance 0.4 ≤ DTI < 0.8, and high drought resistance DTI ≥ 0.8.

[0072] Grain filling stage: During the grain filling stage, corn requires ample water supply, and the drought resistance index threshold can be set higher to screen out varieties that can maintain good growth during the grain filling stage; for example: low drought resistance DTI < 0.7, medium drought resistance 0.7 ≤ DTI < 1.4, and high drought resistance DTI ≥ 1.4. If the leaf area and chlorophyll content of a variety are found to be significantly higher than those of other varieties during the seedling stage, the drought resistance index threshold for that variety can be appropriately increased. If the soil water use efficiency of a variety is found to be high during the grain filling stage, the drought resistance index threshold can be appropriately increased; the details are shown in the following table:

[0073] Seedling stage Jointing stage Heading period Grouting period Low drought resistance DTI<0.5 DTI<0.6 DTI<0.4 DTI<0.7 Moderate drought resistance 0.5≤DTI<1.0 0.6≤DTI<1.2 0.4≤DTI<0.8 0.7≤DTI<1.4 High drought resistance DTI ≥ 1.0 DTI ≥ 1.2 DTI ≥ 0.8 DTI ≥ 1.4

[0074] Table 2. Drought resistance index threshold adjustment table

[0075] The result verification module is configured to verify the effect of the dynamic adjustment threshold by comparing the drought resistance grade classification results before and after the adjustment; based on the verification results, the dynamic adjustment strategy is further optimized to ensure the rationality and accuracy of the drought resistance index threshold.

[0076] The beneficial effects achieved by the above content are: the drought resistance index of corn is calculated through the random forest model, the K-means clustering algorithm is used to divide the drought resistance levels of corn, and a drought resistance index threshold is set for each level; by analyzing the distribution of drought resistance index of corn in different growth stages, an initial drought resistance index threshold is set for each growth stage; and the drought resistance index threshold is dynamically adjusted according to the real-time monitored corn growth data, which can more accurately reflect the drought resistance of corn at different growth stages, so as to formulate a more reasonable drought resistance identification strategy.

[0077] The visualization terminal is configured to display graphs of soil moisture, environmental parameters, and phenotypic parameters in real time at each stage through a user interface, such as a web interface or a mobile application, and generate data comparison graphs between different stages to intuitively display the differences between different stages; for example, generate a comparison graph of chlorophyll content in different growth periods to analyze the growth conditions at different stages; and based on experimental results, generate a visualization graph for drought resistance evaluation to intuitively display the differences in drought resistance of different varieties; for example, generate a bar graph of drought resistance index to compare the drought resistance of different varieties.

[0078] A method for identifying drought resistance of corn comprises the following steps:

[0079] Collect corn plant images from multiple angles and calculate leaf area, plant height, and chlorophyll content to obtain phenotypic parameters;

[0080] Automatically collect soil moisture data at different depths in corn planting areas at regular intervals, and monitor air temperature, humidity, and light intensity in real time;

[0081] Extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, and rank the characteristic variables from high to low according to the characteristic importance score;

[0082] Refer to historical data to classify corn drought resistance levels, and set drought resistance index thresholds for each level based on the different growth stages of corn;

[0083] The random forest model is used to calculate the current drought resistance index of corn. Combined with the current growth period of corn, it is determined whether the drought resistance index is within the drought resistance threshold range and whether the current drought resistance index of corn needs to be adjusted.

[0084] Combined with the current ranking results of corn characteristics importance, field management suggestions are provided for current corn planting.

[0085] Working principle: By collecting corn plant images from multiple angles, soil moisture data at different depths in the corn-planting area are collected, and air temperature, humidity and light intensity are monitored in real time; the random forest model is used to calculate the importance scores of the above-mentioned characteristic variables, and the characteristic vector that contributes most to the prediction of the drought resistance index is identified; corn drought resistance levels are divided, and a drought resistance index threshold is set for each level; the random forest model is used to calculate the current corn drought resistance index, and combined with the current corn growth period, it is determined whether the drought resistance index is within the drought resistance threshold range, and then whether the current corn drought resistance index needs to be adjusted to make it more in line with the actual needs of corn in different growth stages; combined with the current corn feature importance ranking results, field management suggestions are provided for current corn planting, thereby improving the growth performance and yield of corn under drought conditions, and providing scientific guidance for corn drought-resistant breeding and cultivation practices.

[0086] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0087] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A device for identifying drought resistance of corn, characterized in that: The device comprises: a plant phenotype monitoring unit, a soil moisture monitoring unit and a drought resistance identification unit; The plant phenotype monitoring unit is configured to process and analyze corn plant images based on multispectral imaging technology to obtain phenotypic parameters of the corn plants, the phenotypic parameters including leaf area, plant height, and chlorophyll content; The soil moisture monitoring unit is configured to install soil moisture sensors at different depths in the corn planting area, automatically collect soil moisture data at regular intervals, and obtain soil moisture parameters; based on temperature sensors, humidity sensors, and light sensors, it monitors air temperature, humidity, and light intensity in real time; The drought resistance identification unit is configured to integrate phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, calculate the drought resistance index of corn based on a random forest model, and obtain a corn drought resistance identification result.

2. A corn drought resistance identification device according to claim 1, characterized in that: The drought resistance identification unit comprises: A grading module is configured to extract the drought resistance index of all corn varieties or treatments from the prediction results of the random forest model, standardize the drought resistance index, classify the corn drought resistance grades using the K-means clustering algorithm, and set a drought resistance index threshold for each grade; The dynamic adjustment module is configured to dynamically adjust the drought resistance index threshold according to the differences in the growth period of corn.

3. A corn drought resistance identification device according to claim 2, characterized in that: The dynamic adjustment module includes: The strategy formulation module is configured to analyze the distribution of drought resistance index of corn at different growth stages, including minimum, maximum, average, and standard deviation indicators. Based on the physiological characteristics and drought resistance performance of corn at different growth stages, the initial drought resistance index threshold is set for each growth stage, and the drought resistance index threshold is dynamically adjusted based on real-time monitored corn growth data. The result verification module is configured to verify the effect of the dynamic adjustment threshold by comparing the drought resistance grade classification results before and after the adjustment; and further optimize the dynamic adjustment strategy based on the verification results.

4. A corn drought resistance identification device according to claim 2, characterized in that: The drought resistance identification unit further comprises: A feature analysis module is configured to sort the phenotypic parameters of corn plants, soil moisture, temperature and humidity data, and light intensity from high to low according to the feature importance score, and obtain the feature variable that contributes most to the prediction of drought resistance index; The feature ranking module is configured to list the importance scores of all feature variables based on the prediction results of random forests, focusing on analyzing the top-ranked features and explaining their contribution to the drought resistance index; based on the feature importance analysis, it puts forward suggestions for maize drought resistance research and breeding.

5. The corn drought resistance identification device according to claim 2, characterized in that: The drought resistance identification unit further comprises: a feature extraction module configured to collect historical data as training data, including samples with known drought resistance indexes, and extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity from the samples; The data processing module is configured to perform standardization on all characteristic variables to eliminate the effects of different dimensions and orders of magnitude.

6. A corn drought resistance identification device according to claim 2, characterized in that: The drought resistance identification unit further comprises: The model building module is configured to randomly select samples from the training data and randomly select a subset of features each time a node is split. A decision tree is constructed for each sample set until the maximum depth or minimum number of samples is reached. The prediction results of all decision trees are averaged to obtain the final drought resistance index. The model validation module is configured to validate the random forest model using samples that did not participate in the training of each tree, and measure the error between the predicted value and the actual value.

7. The corn drought resistance identification device according to claim 1, characterized in that: The plant phenotype monitoring unit comprises: an image acquisition module configured to use a high-resolution camera to capture images of corn plants from multiple angles, including front, side, and top views; The parameter acquisition module is configured to use an image segmentation algorithm to separate leaves and background for calculating leaf area, measure plant height through a feature point detection algorithm, and calculate chlorophyll content using spectral analysis technology.

8. The corn drought resistance identification device according to claim 1, characterized in that: The soil moisture monitoring unit comprises: a parameter analysis module configured to obtain soil moisture data, temperature and humidity data, and light intensity at different depths, calculate average soil moisture content, daily change rate of soil moisture content, and soil moisture gradient, and calculate the cumulative amount of daily average air temperature, average air humidity, and light intensity; The parameter fusion module is configured to fuse soil moisture data, temperature and humidity data, and light intensity to form comprehensive data.

9. The corn drought resistance identification device according to claim 1, characterized in that: The device further comprises: a visualization terminal; The visualization terminal is configured to display charts of soil moisture, environmental parameters, and phenotypic parameters in real time at each stage through the user interface, generate data comparison charts between different stages, and intuitively display the differences between different stages; and based on the experimental results, generate visualization charts for drought resistance evaluation to intuitively display the differences in drought resistance of different varieties.

10. A method for identifying drought resistance of corn, which is implemented by using the device for identifying drought resistance of corn according to any one of claims 1 to 9, characterized in that: The following steps are involved: Collect corn plant images from multiple angles and calculate leaf area, plant height, and chlorophyll content to obtain phenotypic parameters; Automatically collect soil moisture data at different depths in corn planting areas at regular intervals, and monitor air temperature, humidity, and light intensity in real time; Extract characteristic variables of phenotypic parameters, soil moisture, temperature and humidity data, and light intensity, and rank the characteristic variables from high to low according to the characteristic importance score; Refer to historical data to classify corn drought resistance levels, and set drought resistance index thresholds for each level based on the different growth stages of corn; The random forest model is used to calculate the current drought resistance index of corn. Combined with the current growth period of corn, it is determined whether the drought resistance index is within the drought resistance threshold range and whether the current drought resistance index of corn needs to be adjusted. Combined with the current ranking results of corn characteristic importance, suggestions are provided for current corn planting field management.

Images