Crop growth real-time monitoring system and method

Abstract

The invention discloses a crop growth real-time monitoring system and method, and relates to the technical field of crop monitoring. According to the crop growth real-time monitoring method, crop image data time sequence and soil state time sequence data are continuously acquired in real time, data analysis is performed respectively to obtain an initial agricultural healthy flow index and a soil conditioning index of each monitoring time period, comprehensive analysis is performed to obtain a comprehensive agricultural healthy flow index of each monitoring time period, and prediction analysis is performed to obtain a real-time crop growth monitoring result. According to the method, the comprehensive agricultural healthy flow index of the next monitoring time period is compared and analyzed with the preset comprehensive agricultural healthy flow index interval, and corresponding early warning measures are taken based on the comparison and analysis result, so that the growth state of crops and the change of the soil environment are tracked in real time; therefore, the health state of the crops and the influence of the soil on the growth of the crops are obtained, and the monitoring accuracy and predictability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of crop monitoring, and particularly to a real-time crop growth monitoring system and method. Background Art

[0002] In recent years, the rapid development of sensor technology and wireless communication technology has provided new opportunities for agricultural monitoring. Through a variety of sensors deployed in farmland, environmental data required for crop growth, such as soil humidity, temperature, light intensity, etc., can be obtained in real time, and then the growth status of crops can be monitored. At the same time, with the progress of computer vision and image recognition technology, crop health monitoring based on image data has become a research hotspot. Subtle changes during the crop growth process, such as leaf color, morphology, etc., can provide valuable growth status information.

[0003] The prior art, such as a patent application with publication number CN119205402A, discloses an artificial intelligence-based crop growth monitoring method and system. By obtaining multi-source crop growth data based on sensors, a pre-trained BP crop status prediction model is established based on the BP neural network. The SSA sparrow optimization algorithm is used to optimize the parameters of the BP crop status prediction model. The real-time crop growth data is input into the target BP crop status prediction model for training. Irrigation strategies and fertilization strategies for crops are generated according to the crop growth status, and the irrigation strategies and fertilization strategies are transmitted to the crop irrigation system and fertilization system; the target crop growth data of the crops after irrigation and fertilization is obtained, and the irrigation strategies and fertilization strategies are adjusted according to the target crop growth data. It can provide scientific and reasonable irrigation and fertilization suggestions and management strategies for farmers, thereby optimizing resource allocation and improving crop yield and quality.

[0004] Based on the above scheme, it is found that the limitations of the prior art at least include the following problems. First, the prior art is difficult to reflect the dynamic changes of crop and soil status, and there are limitations in multi-source data fusion and comprehensive analysis, so it is easy to ignore the change trends in different monitoring periods, which in turn affects the prediction accuracy and subsequent decision-making effects. And less consideration is given to the temporal correlation between image data and soil status data, and it is difficult to fully explore the potential laws in the crop growth process, thus easily resulting in relatively limited prediction of the crop growth status. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a real-time crop growth monitoring system and method, which solves the problems that the prior art is difficult to reflect the dynamic changes of crop and soil status and the temporal correlation between image data and soil status data, thus easily affecting the prediction accuracy and subsequent decision-making effects.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for real-time monitoring of crop growth, comprising the following steps: Continuously and real-time obtaining the time series of crop image data and the time series data of soil conditions in a set farmland; Respectively perform data analysis on the time series of crop image data and the time series data of soil conditions in the set farmland to obtain the initial agricultural health flow index and the soil adjustment intelligence index for each monitoring period in the set farmland, and perform comprehensive analysis to obtain the comprehensive agricultural health flow index for each monitoring period in the set farmland; Perform predictive analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland; Compare and analyze the comprehensive agricultural health flow index for the next monitoring period in the set farmland with the preset comprehensive agricultural health flow index range, and take corresponding warning measures based on the comparison and analysis results; Among them, the specific formula for calculating the comprehensive agricultural health flow index for each monitoring period in the set farmland is as follows: ; Wherein, ZhW i is the comprehensive agricultural health flow index for the i-th monitoring period in the set farmland, CzH i , TrX i are successively the initial agricultural health flow index and the soil adjustment intelligence index for the i-th monitoring period in the set farmland, α 1 , β 1 , α 2 , β 2 , β 3 are successively the initial agricultural health coefficient, the initial agricultural health adjustment coefficient, the soil adjustment intelligence index, the soil adjustment intelligence coefficient, the soil adjustment intelligence adjustment coefficient, and the interaction coefficient stored in the database, α 1 +α 2 =1, i = 1, 2, 3,..., i 0 , i 0 is the number of monitoring periods.

[0007] Further, the time series of crop image data includes the pixel value and two-dimensional coordinates of each pixel point for each monitoring period, and the time series data of soil conditions includes the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value for each monitoring period.

[0008] Further, the specific steps to obtain the initial agricultural health flow index for each monitoring period in the set farmland are as follows: Identify and analyze the pixel values and two-dimensional coordinates of each pixel point in each monitoring period in the set farmland to obtain several leaf areas and leaf stem areas in each monitoring period in the set farmland; Comprehensively analyze the pixel values of each pixel point in each leaf area in each monitoring period in the set farmland to obtain the leaf green saturation index in each monitoring period in the set farmland; Read the two-dimensional coordinates of each leaf edge pixel point in each leaf area in each monitoring period in the set farmland and conduct comprehensive analysis to obtain the leaf structure diversity index in each monitoring period in the set farmland; Read the pixel values and two-dimensional coordinates of each leaf stem pixel point in each leaf stem area in each monitoring period in the set farmland and conduct comprehensive analysis to obtain the leaf stem curve index in each monitoring period in the set farmland; Comprehensively analyze the leaf green saturation index, leaf structure diversity index, and leaf stem curve index in each monitoring period in the set farmland to obtain the initial agricultural health flow index in each monitoring period in the set farmland.

[0009] Further, the specific steps to obtain several leaf areas and leaf stem areas in each monitoring period in the set farmland are as follows: Perform color space conversion processing on the pixel values of each pixel point in each monitoring period in the set farmland to obtain the hue value, saturation value, and lightness value of each pixel point in each monitoring period in the set farmland, and conduct threshold segmentation processing to obtain several color prediction leaf areas and color prediction leaf stem areas in each monitoring period in the set farmland; And perform grayscale processing on the pixel values of each pixel point in each monitoring period in the set farmland; Based on the pixel values of each pixel point in each monitoring period in the set farmland after grayscale processing, conduct edge detection processing to obtain several edge prediction leaf areas and edge prediction leaf stem areas in each monitoring period in the set farmland; Respectively read the two-dimensional coordinates of each color prediction leaf area, color prediction leaf stem area, edge prediction leaf area, and edge prediction leaf stem area in each monitoring period in the set farmland and conduct comprehensive analysis to obtain several leaf areas and leaf stem areas in each monitoring period in the set farmland.

[0010] Further, the specific formula for calculating the initial agricultural health flow index for each monitoring period in the set farmland is as follows: ; where CzH i is the initial agricultural health flow index for the i-th monitoring period in the set farmland, YbH i , YdX i , QwZ i are the leaf green saturation index, leaf structure diversity index, and leaf stem curve index for the i-th monitoring period in the set farmland in sequence, Φ 1 , Φ 2 , Φ3 、 Φ 4 、 Φ 5 are the green saturation adjustment coefficient, structure diversity adjustment coefficient, difference adjustment coefficient, curved pattern adjustment coefficient, and ratio adjustment coefficient stored in the database in sequence, where \(i = 1, 2, 3,\cdots, i\) 0 , \(i\) 0 is the number of monitoring periods.

[0011] Furthermore, the specific steps to obtain the leaf-stem curved pattern index of each leaf-stem area in each monitoring period in the set farmland are as follows: comprehensively analyze the pixel values of each leaf-stem pixel point in each leaf-stem area in each monitoring period in the set farmland to obtain the leaf-stem texture complexity index of each leaf-stem area in each monitoring period in the set farmland; read the two-dimensional coordinates of each leaf-stem edge pixel point in each leaf-stem area in each monitoring period in the set farmland and conduct comprehensive analysis to obtain the leaf-stem morphological tortuosity index of each leaf-stem area in each monitoring period in the set farmland; comprehensively analyze the leaf-stem texture complexity index and the leaf-stem morphological tortuosity index of each leaf-stem area in each monitoring period in the set farmland to obtain the leaf-stem curved pattern index of each monitoring period in the set farmland.

[0012] Furthermore, the specific steps to obtain the soil intelligence adjustment index of each monitoring period in the set farmland are as follows: obtain the soil state reference set in the set farmland, where the soil state reference set includes the soil suitability reference index, soil compaction reference value, soil humic acid content reference value, soluble nutrient reference index, and soil microbial activity reference value; comprehensively analyze the soil state reference set in the set farmland and the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value of each monitoring period to obtain the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland; and comprehensively analyze the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland to obtain the soil intelligence adjustment index of each monitoring period in the set farmland.

[0013] Furthermore, the specific formulas for calculating the soil suitability difference index and the soil intelligence adjustment index of each monitoring period in the set farmland are as follows: ; where \(TsC\) i is the soil suitability difference index of the \(i\)-th monitoring period in the set farmland, \(TsY\) i is the soil suitability index of the \(i\)-th monitoring period in the set farmland, \(CsY\) is the soil suitability reference index in the set farmland, \(TrX\) iTo set the soil adjustment intelligence index, KyC, for the i-th monitoring period in the farmland i , DhC i , XkC i , YwC i are, in sequence, the soil compaction degree difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index for the i-th monitoring period in the set farmland. η 1 , η 2 , η 3 , η 4 , η 5 , η 6 are, in sequence, the soil suitability adjustment coefficient, soil compaction adjustment coefficient, humic acid adjustment coefficient, humic acid adjustment coefficient, microbial activity adjustment coefficient, and soil interaction coefficient stored in the database. i = 1, 2, 3,..., i 0 , i 0 is the number of monitoring periods.

[0014] Furthermore, the specific steps to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland are as follows: Conduct trend analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland to obtain several groups of comprehensive agricultural health flow index change rates in the set farmland; and conduct mean analysis on the several groups of comprehensive agricultural health flow index change rates in the set farmland to obtain the mean value of the comprehensive agricultural health flow index change rates in the set farmland, and conduct comprehensive analysis to obtain the comprehensive agricultural health flow prediction index in the set farmland, and regard it as the comprehensive agricultural health flow prediction index for the next monitoring period in the set farmland.

[0015] A real-time monitoring system for crop growth includes: a data acquisition module, an image analysis module, a data analysis module, a prediction analysis module, and a growth warning module; the data acquisition module is used to continuously and real-time acquire the time series of crop image data and the time series data of soil conditions in the set farmland; the image analysis module is used to conduct image analysis on the time series of crop image data in the set farmland to obtain the initial agricultural health flow index for each monitoring period in the set farmland; the data analysis module is used to conduct data analysis on the time series of crop image data in the set farmland to obtain the soil adjustment intelligence index for each monitoring period in the set farmland, and conduct comprehensive analysis in combination with the initial agricultural health flow index to obtain the comprehensive agricultural health flow index for each monitoring period in the set farmland; the prediction analysis module is used to conduct prediction analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland; the growth warning module is used to compare and analyze the comprehensive agricultural health flow index for the next monitoring period in the set farmland with a preset comprehensive agricultural health flow index threshold, and take corresponding warning measures based on the comparison and analysis results.

[0016] The present invention has the following beneficial effects: (1) The real-time monitoring method for crop growth can continuously and real-time obtain the time series of crop image data and the time series of soil state data, so as to be able to track the growth state of crops and the changes in soil environment in real time, thereby improving the timeliness of monitoring, ensuring that problems in the crop growth process can be discovered and intervened in a timely manner, and through the comprehensive analysis of image data and soil state data, the health state of crops and the impact of soil on their growth can be obtained, and then provide a strong basis for formulating precise agricultural management measures, and then improve the accuracy and predictability of monitoring.

[0017] (2) The real-time monitoring method for crop growth can predict and analyze the comprehensive agricultural health flow index for each monitoring period, so as to predict in advance the growth state of crops and the changes in soil conditions, and then understand the future growth trend of crops. For example, when the comprehensive agricultural health flow index is predicted to show an abnormal decline in the next monitoring period, farmers are notified in advance to adjust farm management measures, such as increasing the irrigation amount or adjusting the fertilization plan, so as to effectively avoid crop damage caused by environmental changes. When there are sudden climate changes or soil fertility changes, preventive measures are taken in a timely manner to improve the growth efficiency and yield of crops.

[0018] (3) The real-time monitoring method for crop growth can compare the predicted comprehensive agricultural health flow index of the next monitoring period with a preset interval, so as to achieve a dynamic warning mechanism. When the growth state of crops or soil conditions deviate from the normal range, an alarm can be automatically triggered and corresponding measures can be taken, thus avoiding crop losses caused by lagging or incorrect management measures, and automatically adjusting agricultural operations according to the warning results, thereby improving the adaptability of agricultural production and enhancing the risk control effect of agricultural management, and ensuring the high efficiency and sustainable development of farmland management.

[0019] (4) The real-time monitoring system for crop growth realizes the precision of farmland management through the collaborative work of multiple modules. By real-time obtaining and analyzing farmland data, the system can provide more scientific and timely management suggestions for farmers according to the growth trend of crops and soil impact. For example, when the system predicts that drought or nutrient deficiency is about to occur, corresponding preventive measures can be taken in advance, such as automatically adjusting the irrigation amount or fertilization amount, thereby reducing human errors in farmland management and avoiding over-irrigation or over-fertilization, and then ensuring the reasonable use of resources. At the same time, the system can timely identify and respond to potential growth problems, and remind farmers to take quick response measures through the warning mechanism, so as to effectively reduce crop losses and improve the efficiency and benefit of farmland management.

[0020] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. Brief Description of the Drawings

[0021] Figure 1 This is a flowchart of a method for real-time monitoring of crop growth according to the present invention.

[0022] Figure 2 This is a flowchart of the steps for obtaining the initial agricultural health flow index for each monitoring period in a set farmland in a method for real-time monitoring of crop growth according to the present invention.

[0023] Figure 3 This is a block diagram of a system for real-time monitoring of crop growth according to the present invention. Detailed Embodiment

[0024] Please refer to Figure 1 , an embodiment of the present invention provides a technical solution: a method for real-time monitoring of crop growth, including the following steps: continuously and real-time obtaining the time series of crop image data and the time series data of soil conditions in a set farmland; respectively performing data analysis on the time series of crop image data and the time series data of soil conditions in the set farmland to obtain the initial agricultural health flow index (i.e., measuring the health status of crop growth) and the soil adjustment intelligence index (i.e., measuring the impact of the soil conditions where the crops are located on their growth status) for each monitoring period in the set farmland, and performing comprehensive analysis to obtain the comprehensive agricultural health flow index for each monitoring period in the set farmland; performing predictive analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland; comparing and analyzing the comprehensive agricultural health flow index for the next monitoring period in the set farmland with a preset comprehensive agricultural health flow index range, and taking corresponding warning measures based on the comparison and analysis results; wherein, the specific formula for calculating the comprehensive agricultural health flow index for each monitoring period in the set farmland is as follows: ; where, ZhW i is the comprehensive agricultural health flow index for the i-th monitoring period in the set farmland, CzH i is the initial agricultural health flow index for the i-th monitoring period in the set farmland, α 1 is the initial agricultural health coefficient stored in the database, β 1 is the initial agricultural health adjustment coefficient stored in the database, TrX i is the soil adjustment intelligence index for the i-th monitoring period in the set farmland, α 2 is the soil adjustment intelligence coefficient stored in the database, β 2 is the soil adjustment intelligence adjustment coefficient stored in the database, β 3 is the interaction coefficient stored in the database, α 1 +α 2 =1, i = 1, 2, 3,..., i 0 , i 0 is the number of monitoring periods.

[0025] It should be noted that in the formula is used to adjust the superposition effect of the initial agricultural health flow index and the soil intelligence adjustment index, and avoid the comprehensive agricultural health flow index being too high or too low.

[0026] α 1 and α 2 can be obtained through the following steps: Read the initial agricultural health flow index and the soil intelligence adjustment index of each monitoring period in the set farmland, conduct mean analysis to obtain the mean value of the initial agricultural health flow index and the mean value of the soil intelligence adjustment index in the set farmland, and conduct summation analysis to obtain the growth health sum value. Then, conduct ratio analysis on the mean value of the initial agricultural health flow index and the mean value of the soil intelligence adjustment index in the set farmland respectively with the growth health sum value, and use the ratio analysis results as the corresponding coefficients.

[0027] β 1 and β 2 and β 3 can be obtained through the following steps: Use historical data, combine the initial agricultural health flow index and the soil intelligence adjustment index indicators, conduct statistical regression analysis to quantify the specific impact of each factor on the comprehensive agricultural health flow index, so as to fit the initial weight value. Secondly, use the sensitivity analysis method to adjust the value range of each coefficient, observe its impact on the evaluation result of the comprehensive agricultural health flow index, ensure the stability and rationality of the model, and based on the regional characteristics and actual situation, correct and optimize the initially fitted coefficients, and finally determine the coefficient values applicable to a specific region.

[0028] The specific steps for taking corresponding early warning measures based on the comparison analysis results are as follows: If the comprehensive agricultural health flow index of the next monitoring period in the set farmland is lower than the lower limit of the preset comprehensive agricultural health flow index range (i.e., the minimum value of the range), then take the first early warning measure (i.e., provide relevant personnel with suggestions for restorative irrigation. If the soil moisture is insufficient, appropriate irrigation should be carried out, but over-irrigation should be avoided to restore soil moisture; apply appropriate fertilizers: apply organic fertilizers or slow-release fertilizers to supplement the nutrients in the soil, especially nitrogen, phosphorus, potassium and other nutrient elements, but the amount of fertilizer application should be controlled to prevent other problems caused by over-fertilization; improve the soil structure. If the soil is too compact, measures such as loosening the soil and adding organic matter can be taken to improve the soil permeability and water retention capacity; strengthen pest and disease control, increase the monitoring of pests and diseases, and prevent their further deterioration in a timely manner); If the comprehensive agricultural health flow index of the next monitoring period in the set farmland is within the preset comprehensive agricultural health flow index range, then take the second early warning measure (i.e., provide relevant personnel with suggestions for normal management, continue with regular irrigation, fertilization and soil management to maintain the normal growth environment of the crops); If the comprehensive agricultural health flow index of the next monitoring period in the farmland is set higher than the preset upper limit of the comprehensive agricultural health flow index interval (i.e. the maximum value of the interval), the third warning measure will be taken (i.e. providing relevant personnel with suggestions on reducing irrigation and irrigation amount to prevent excessive water from causing the soil to be too wet, affecting the root system's breathing and water absorption capacity; reducing fertilization and reducing fertilization intensity to avoid excessive nutrient supply leading to excessive crop growth or excess nutrition; increasing soil permeability by loosening the soil or adding organic matter to increase soil permeability to avoid root problems caused by excessive moisture or excess nutrients in the soil).

[0029] The crop image data time series includes the pixel value and two-dimensional coordinates (including X-axis coordinate value and Y-axis coordinate value) of each pixel point in each monitoring period. The soil status time series data includes the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value in each monitoring period.

[0030] Among them, the soil suitability index is a measure of the suitability of the soil state for the healthy growth of crops. It obtains the soil moisture value (soil moisture sensor), soil temperature value (soil temperature sensor), soil pH value (pH value measuring instrument), and soil conductivity value (conductivity sensor), and performs standardization. Based on the standardized processing results, weighted processing is performed. The result is this parameter, which is used to reflect whether an environment suitable for plant growth is provided to ensure the healthy growth of plants and nutrient absorption.

[0031] The soil compaction value is the ratio between the mass of soil particles and the soil volume, indicating the compactness of the soil. It can be obtained through a soil compaction sensor and is used to reflect the development of plant roots and the water and air circulation.

[0032] The soil humic acid content value is the product of the decomposition of soil organic matter. It can be obtained by a near-infrared spectrometer (that is, the soil's reflectance spectrum in the near-infrared region is used to evaluate its organic matter content, and by establishing a standard curve, the humic acid content in the soil is inferred). It is used to reflect the improvement of soil fertility, enhance the efficiency of plant nutrient utilization, and thus promote healthy plant growth.

[0033] The soluble nutrient index is a comprehensive indicator for measuring the nitrogen, phosphorus and potassium content in the soil for soil fertility and plant growth. The nitrogen, phosphorus and potassium content in the soil can be measured by portable soil testing instruments, and the measurement results are uploaded to the database. The nitrogen content, phosphorus content and potassium content in the soil are weighted, and the result is the parameter, which is used to promote plant roots to absorb nutrients, enhance soil fertility, and help plants grow healthily.

[0034] The soil microbial activity value reflects the microbial activity in the soil and affects soil fertility. Here, it is represented by the respiration rate of carbon dioxide (which can be measured by a soil respiration meter and the measurement results are uploaded to the database). It can be obtained by acquiring the respiration rates of carbon dioxide at multiple soil monitoring points and performing an average treatment. The resulting value is this parameter, which is used to reflect the activity level of microorganisms in the soil, thereby indirectly affecting the nutrient absorption and growth of plants.

[0035] Specifically, as Figure 2 shown, the specific steps to obtain the initial agricultural health flow index for each monitoring period in the set farmland are as follows: Identify and analyze the pixel values and two-dimensional coordinates of each pixel point for each monitoring period in the set farmland to obtain several leaf areas and leaf stem areas for each monitoring period in the set farmland; Comprehensively analyze the pixel values of each pixel point in each leaf area for each monitoring period in the set farmland (i.e., perform color space conversion, generate the green saturation value of each pixel point in each leaf area, and perform an average treatment) to obtain the leaf green saturation index for each monitoring period in the set farmland; Read the two-dimensional coordinates of each leaf edge pixel point in each leaf area for each monitoring period in the set farmland and perform a comprehensive analysis to obtain the leaf structure diversity index for each monitoring period in the set farmland; Read the pixel values and two-dimensional coordinates of each leaf stem pixel point in each leaf stem area for each monitoring period in the set farmland and perform a comprehensive analysis to obtain the leaf stem curve index for each monitoring period in the set farmland; Comprehensively analyze the leaf green saturation index, leaf structure diversity index, and leaf stem curve index for each monitoring period in the set farmland to obtain the initial agricultural health flow index for each monitoring period in the set farmland.

[0036] And the specific steps to obtain the leaf structure diversity index for each monitoring period in the set farmland are as follows: Read the two-dimensional coordinates of each leaf edge pixel point in each leaf area for each monitoring period in the set farmland and perform a comprehensive analysis (i.e., analyze based on the polygon area formula) to obtain the area value of each leaf area for each monitoring period in the set farmland; And comprehensively analyze the two-dimensional coordinates of each leaf edge pixel point in each leaf area for each monitoring period in the set farmland (i.e., based on the curvature formula) to obtain the edge complexity index of each leaf area for each monitoring period in the set farmland, and comprehensively analyze it in combination with the area value (i.e., perform weighted treatment and then average treatment) to obtain the leaf structure diversity index for each monitoring period in the set farmland.

[0037] The specific formula for calculating the initial agricultural health flow index for each monitoring period in the set farmland is as follows: ; where CzH i is the initial agricultural health flow index for the i-th monitoring period in the set farmland, YbHi To set the leaf green saturation index for the i-th monitoring period in the farmland, Φ 1 is the green saturation adjustment coefficient stored in the database, YdX i To set the leaf structure diversity index for the i-th monitoring period in the farmland, Φ 2 is the structure diversity adjustment coefficient stored in the database, Φ 3 is the difference adjustment coefficient stored in the database, QwZ i To set the leaf-stem texture index for the i-th monitoring period in the farmland, Φ 4 is the texture adjustment coefficient stored in the database, Φ 5 is the ratio adjustment coefficient stored in the database, i = 1, 2, 3, …, i 0 , i 0 is the number of monitoring periods.

[0038] It should be explained that is used to adjust the difference between the leaf green saturation index and the leaf structure diversity index, preventing abnormal fluctuations in the initial agricultural health flow index caused by excessive differences.

[0039] In the formula is used to adjust the ratio difference between the leaf green saturation index and the leaf structure diversity index, preventing extreme effects on the initial agricultural health flow index caused by excessive differences.

[0040] Φ 1 、Φ 2 、Φ 3 、Φ 4 、Φ 5 can be obtained through the following steps: Based on historical data, determine the initial influence weights of each variable (such as leaf green saturation index, leaf structure diversity index, leaf-stem texture index) on the initial agricultural health flow index through statistical regression analysis. Then, use the sensitivity analysis method to adjust the value range of the coefficients to evaluate the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as multi-objective optimization) to ensure that the formula can accurately reflect the health status of actual crops. Fine-tune the coefficients based on different regional characteristics to ensure its applicability to specific initial agricultural health flow assessment requirements.

[0041] In this implementation plan, through the comprehensive analysis of multiple detailed indicators such as the green saturation, structural diversity, and leaf-stem texture of crop leaves, the growth status of crops can be comprehensively and accurately evaluated. These indices reflect the growth quality and environmental adaptability of crops. Through precise image analysis, changes in leaf health, leaf-stem structure, etc. can be carefully captured, which helps to detect growth abnormalities earlier, such as pest and disease problems, nutrient deficiencies, etc., and then provides a basis for timely adjustment of management measures (such as fertilization, irrigation, etc.). Secondly, through various adjustment coefficients in the formula, the unreasonable fluctuations of the initial agricultural health flow index caused by differences between different indicators can be effectively avoided, thus ensuring the stability and accuracy of the agricultural health flow index, and then enhancing the reliability of the monitoring results. Finally, through methods such as regression analysis, sensitivity analysis, and multi-objective optimization based on historical data, the coefficients are adjusted to ensure that the formula can accurately reflect the actual health status of crops, thereby improving the calculation accuracy and being able to dynamically fine-tune the coefficients according to different regional characteristics, so as to flexibly adapt to various crops and agricultural environments, and then promote the precision and intelligence of agricultural management.

[0042] Specifically, the specific steps to obtain several leaf regions and leaf-stem regions for each monitoring period in the set farmland are as follows: perform color space conversion processing (i.e., convert to HSV color space) on the pixel values of each pixel point for each monitoring period in the set farmland to obtain the hue value, saturation value, and lightness value of each pixel point for each monitoring period in the set farmland, and perform threshold segmentation processing to obtain several color-predicted leaf regions and color-predicted leaf-stem regions for each monitoring period in the set farmland; and perform grayscale processing on the pixel values of each pixel point for each monitoring period in the set farmland; perform edge detection processing based on the pixel values of each pixel point for each monitoring period in the set farmland after grayscale processing to obtain several edge-predicted leaf regions and edge-predicted leaf-stem regions for each monitoring period in the set farmland; respectively read the two-dimensional coordinates of each color-predicted leaf region, color-predicted leaf-stem region, edge-predicted leaf region, and edge-predicted leaf-stem region for each monitoring period in the set farmland for comprehensive analysis (i.e., in each color-predicted leaf region and the corresponding edge-predicted leaf region, the pixel points with the same two-dimensional coordinates are merged, and those with different coordinates are discarded to obtain the leaf region, and the leaf-stem region is obtained in the same way), to obtain several leaf regions and leaf-stem regions for each monitoring period in the set farmland.

[0043] Among them, the specific steps of the threshold segmentation processing are as follows: analyze the hue value, saturation value, and lightness value of each pixel point for each monitoring period in the set farmland respectively with the set first hue interval, first saturation threshold, first lightness threshold, second hue interval, second saturation threshold, and second lightness threshold; If the hue of each pixel point in each monitoring period within the farmland is within the first hue interval, and the saturation is higher than the first saturation threshold, and the lightness is higher than the first lightness threshold, then mark this pixel point as a leaf pixel point; If the hue of each pixel point in each monitoring period within the farmland is within the second hue interval, and the saturation is higher than the second saturation threshold, and the lightness is higher than the second lightness threshold, then mark this pixel point as a leaf stalk pixel point.

[0044] And the first hue interval is higher than the second hue interval, the first saturation threshold is higher than the second saturation threshold, and the first lightness threshold is higher than the second lightness threshold.

[0045] The edge detection process uses the Canny edge detection. First, Gaussian filtering is performed for denoising, then the gradient of the image is calculated and edge detection is carried out, and finally non-maximum suppression and double-threshold processing are used to accurately locate the edges.

[0046] In this implementation scheme, through the color space conversion (HSV color space) and threshold segmentation processing of the farmland image, the leaf and leaf stalk areas can be efficiently and accurately distinguished. And the comprehensive analysis of the hue, saturation and lightness values can automatically identify the leaf and leaf stalk areas according to different color characteristics, thus avoiding the errors of traditional manual identification methods, and providing a reliable data basis for subsequent health assessment and monitoring. Secondly, through the edge detection processing of the grayscale image and using the Canny edge detection algorithm, combined with Gaussian filtering, gradient calculation, non-maximum suppression and double-threshold processing, the edges of the crop leaves and leaf stalks can be accurately located, so that the boundaries of the crops can be clearly identified, improving the accuracy of subsequent analysis. Finally, by comprehensively analyzing the two-dimensional coordinates of the color prediction and edge prediction areas, the pixel points with the same coordinates can be efficiently merged, and the irrelevant data can be removed, thus improving the accuracy and processing efficiency of the data, which helps to reduce errors and ensure the accuracy of the results, and then reflects the real growth status of the crops, and provides more accurate information support for subsequent crop health monitoring and management decisions.

[0047] Specifically, the specific steps to obtain the leaf and stem curve index of each leaf and stem area in each monitoring period in the set farmland are as follows: comprehensively analyze the pixel values of each leaf and stem pixel in each leaf and stem area in each monitoring period in the set farmland (that is, first perform grayscale processing to obtain the grayscale pixel values of each pixel in each leaf and stem area in each monitoring period in the set farmland, and analyze based on the local binary pattern and the grayscale pixel values of each neighborhood pixel in the set neighborhood) to obtain the leaf and stem texture complexity index of each leaf and stem area in each monitoring period in the set farmland; read the two-dimensional coordinates of each leaf and stem edge pixel in each leaf and stem area in each monitoring period in the set farmland, and perform comprehensive analysis to obtain the leaf and stem morphological tortuosity index of each leaf and stem area in each monitoring period in the set farmland; comprehensively analyze the leaf and stem texture complexity index and the leaf and stem morphological tortuosity index of each leaf and stem area in each monitoring period in the set farmland (that is, perform weighted processing and then perform mean processing) to obtain the leaf and stem curve index of each monitoring period in the set farmland.

[0048] Among them, the specific formula for calculating the leaf and stem morphological tortuosity index of each leaf and stem area in each monitoring period in the set farmland is as follows: ; where XqT ij is the leaf and stem morphological tortuosity index of the j-th leaf and stem area in the i-th monitoring period in the set farmland, (X ij（m+1） , Y ij（m+1） ) is the two-dimensional coordinate of the (m + 1)-th leaf and stem edge pixel of the j-th leaf and stem area in the i-th monitoring period in the set farmland, (X ijm , Y ijm ) is the two-dimensional coordinate of the m-th leaf and stem edge pixel of the j-th leaf and stem area in the i-th monitoring period in the set farmland, (X ij（m-1） , Y ij（m-1） ) is the two-dimensional coordinate of the m-th leaf and stem edge pixel of the j-th leaf and stem area in the i-th monitoring period in the set farmland, (X ij1 , Y ij1 ) is the two-dimensional coordinate of the 1-st leaf and stem edge pixel of the j-th leaf and stem area in the i-th monitoring period in the set farmland, i = 1, 2, 3,..., i 0 , i 0 is the number of monitoring periods, j = 1, 2, 3,..., j 0 , j 0 is the number of leaf and stem areas, m = 1, 2, 3,..., m 0 , m 0 is the number of leaf and stem areas.

[0049] In this embodiment, by grayscale processing of the leaf stem pixel value and combining it with local binary pattern analysis, the texture complexity of the leaf stem area can be accurately extracted and the structural characteristics of the leaf stem can be reflected, so that potential problems of the leaf stem, such as disease infection or malnutrition, can be discovered in time, thereby providing early warnings for farmland management. Secondly, the leaf stem morphological tortuosity index is calculated, which can reflect the growth state of the leaf stem, especially the degree of distortion of its growth morphology. Normal leaf stems often present a smooth curve, while when affected by pressure, environment or disease, leaf stems are prone to twisting, bending and other changes. Through accurate two-dimensional coordinate analysis and calculation of tortuosity index, the growth morphology of the leaf stem can be deeply understood, and possible growth obstacles, such as lack of water or mechanical damage, can be identified in time. Finally, by weighting and averaging the leaf stem texture complexity index and the leaf stem morphological tortuosity index, multiple factors can be comprehensively considered to obtain a more accurate leaf stem tortuosity index, thereby improving the comprehensive assessment of the health status of the leaf stem, so that farmers can have a more comprehensive understanding of the growth of the leaf stem, and then formulate more effective irrigation, fertilization and prevention and control measures, thereby improving the health management level of crops.

[0050] Specifically, the specific steps for obtaining the soil intelligence index of each monitoring period in the set farmland are as follows: obtain a soil state reference set in the set farmland, the soil state reference set including a soil suitability reference index, a soil compaction reference value, a soil humic acid content reference value, a soluble nutrient reference index, and a soil microbial activity reference value; comprehensively analyze the soil state reference set in the set farmland and the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value of each monitoring period to obtain the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland; and comprehensively analyze the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland to obtain the soil intelligence index of each monitoring period in the set farmland.

[0051] Among them, the soil suitability reference index is obtained by obtaining the soil moisture reference value, soil temperature reference value, soil pH reference value, and soil conductivity reference value, and performing standardization, and then performing weighted processing based on the standardized processing result. The result is the parameter, and the soil moisture reference value, soil temperature reference value, soil pH reference value, and soil conductivity reference value can all be obtained through regional agricultural databases.

[0052] Reference values ​​for soil compaction, soil humic acid content, and soil microbial activity can all be obtained through regional agricultural databases.

[0053] The soluble nutrient reference index obtains the reference nitrogen content, reference phosphorus content, and reference potassium content in the soil and performs a weighting process. The resulting value is this parameter, and the reference nitrogen content, reference phosphorus content, and reference potassium content in the soil can all be obtained from the regional agricultural database.

[0054] The specific formulas for calculating the soil suitability difference index and the soil intelligence adjustment index for each monitoring period in the set farmland are as follows: ; where TsC i is the soil suitability difference index for the i-th monitoring period in the set farmland, TsY i is the soil suitability index for the i-th monitoring period in the set farmland, CsY is the soil suitability reference index for the set farmland, TrX i is the soil intelligence adjustment index for the i-th monitoring period in the set farmland, η 1 is the soil suitability adjustment coefficient stored in the database, KyC i is the soil compaction difference index for the i-th monitoring period in the set farmland, η 2 is the soil compaction adjustment coefficient stored in the database, DhC i is the soil humic acid content difference index for the i-th monitoring period in the set farmland, η 3 is the humic acid adjustment coefficient stored in the database, XkC i is the soluble nutrient difference index for the i-th monitoring period in the set farmland, η 4 is the humic acid adjustment coefficient stored in the database, YwC i is the soil microbial activity difference index for the i-th monitoring period in the set farmland, η 5 is the microbial activity adjustment coefficient stored in the database, η 6 is the soil interaction coefficient stored in the database, i = 1, 2, 3,..., i 0 , i 0 is the number of monitoring periods.

[0055] It should be explained that the calculation logics of the soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index for each monitoring period in the set farmland are the same as that of the soil suitability difference index.

[0056] In the formula is used to adjust the superposition effect of the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index, to avoid the soil intelligence adjustment index being too high or too low.

[0057] η 1 、η 2 、η3 、 η 4 、 η 5 can be obtained through the following steps: Using historical data, evaluate the influence degree of each variable (soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, soil microbial activity difference index) on the soil intelligence adjustment index through statistical modeling and regression analysis, so as to fit the initial weight value. Then, based on sensitivity analysis, adjust the value range of these coefficients to ensure that the formula has good adaptability to soil changes under different environmental conditions. Next, use scenario simulation technology (such as the prediction model of soil intelligence adjustment index in different regions) to further optimize the applicability of the coefficients, and reasonably correct the weight coefficients for specific regions.

[0058] The specific implementation example of calculating the soil intelligence adjustment index of the first monitoring period in the set farmland is as follows. The existing data are as follows: The soil suitability index of the first monitoring period in the set farmland is: 0.79.

[0059] The soil compaction value of the first monitoring period in the set farmland (unit: g / cm³) is: 1.45.

[0060] The soil humic acid content value of the first monitoring period in the set farmland is: 0.16.

[0061] The soluble nutrient index of the first monitoring period in the set farmland is: 0.17.

[0062] The soil microbial activity value of the first monitoring period in the set farmland (unit: µmol / m²·s) is: 0.10.

[0063] The soil suitability reference index in the set farmland is: 0.72.

[0064] The soil compaction reference value in the set farmland (unit: g / cm³) is: 1.30.

[0065] The soil humic acid content reference value in the set farmland is: 0.12.

[0066] The soluble nutrient reference index in the set farmland is: 0.20.

[0067] The soil microbial activity reference value in the set farmland (unit: µmol / m²·s) is: 0.12.

[0068] Substitute the above data into the specific formulas for calculating the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland, and obtain: Set the soil suitability difference index for the first monitoring period in the farmland to approximately: 0.10.

[0069] Set the soil compaction difference index for the first monitoring period in the farmland to approximately: 0.12.

[0070] Set the soil humic acid content difference index for the first monitoring period in the farmland to approximately: 0.33.

[0071] Set the soluble nutrient difference index for the first monitoring period in the farmland to approximately: 0.15.

[0072] Set the soil microbial activity difference index for the first monitoring period in the farmland to approximately: 0.17.

[0073] The soil suitability adjustment coefficient stored in the database is approximately: 0.25.

[0074] The soil compaction adjustment coefficient stored in the database is approximately: 0.28.

[0075] The humic acid adjustment coefficient stored in the database is approximately: 0.31.

[0076] The humic acid adjustment coefficient stored in the database is approximately: 0.23.

[0077] The microbial activity adjustment coefficient stored in the database is approximately: 0.21.

[0078] The soil interaction coefficient stored in the database is approximately: 1.73.

[0079] And substitute the above data into the specific formula for calculating the soil adjustment intelligence index for each monitoring period in the farmland, and obtain:

[0080] In this implementation plan, through the comprehensive analysis of the soil status reference set and monitoring data, the multi-dimensional health status of the soil can be deeply evaluated, accurately reflecting the overall health status of the soil, thereby helping farmers understand whether the soil is in the best state and promptly detecting potential soil problems. Secondly, through the calculation of the soil intelligence adjustment index, the soil management measures can be adjusted in real time according to the soil status differences in each monitoring period. This process includes the comprehensive analysis of the difference indexes of various soil parameters and the application of adjustment coefficients to ensure that the soil intelligence adjustment index reflects the actual situation of soil changes, thus significantly improving the refined and precise management level of agricultural production. Finally, by combining historical data and regression analysis to optimize the calculation formula of the soil intelligence adjustment index, its adaptability under different soil conditions can be ensured, and by correcting the weight coefficients for specific regions, the precision of soil management can be maximized, thereby avoiding management mistakes caused by regional differences and ensuring that the soil management strategy has higher practical application value and operability.

[0081] Specifically, the specific steps to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland are as follows: Conduct a trend analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland (that is, the difference result of subtracting the comprehensive agricultural health flow index of the previous period from that of the next period, divided by the period time), to obtain several groups of comprehensive agricultural health flow index change rates in the set farmland; and conduct a mean analysis on the several groups of comprehensive agricultural health flow index change rates in the set farmland to obtain the mean value of the comprehensive agricultural health flow index change rates in the set farmland, and conduct a comprehensive analysis to obtain the comprehensive agricultural health flow prediction index in the set farmland, which is regarded as the comprehensive agricultural health flow prediction index for the next monitoring period in the set farmland.

[0082] Among them, the specific formula for calculating the comprehensive agricultural health flow prediction index in the set farmland is as follows: ; where ZhY is the comprehensive agricultural health flow prediction index in the set farmland, ZhW i is the comprehensive agricultural health flow index for the i-th monitoring period in the set farmland, BwH n is the n-th group of comprehensive agricultural health flow index change rates in the set farmland, μ n1 is the weighting coefficient of the n-th group of comprehensive agricultural health flow index change rates in the set area, is the first change adjustment coefficient stored in the database, BwH n+1 is the (n + 1)-th group of comprehensive agricultural health flow index change rates in the set farmland, μ n2 is the weighting coefficient of the (n + 1)-th group of comprehensive agricultural health flow index change rates in the set area, is the second change adjustment coefficient stored in the database, is the interactive change adjustment coefficient stored in the database, i = 1, 2, 3,..., i 0 , i0 is the number of monitoring periods, n = 1, 2, 3, …, n 0 , n 0 is the number of groups of the change rate of the comprehensive agricultural health flow index, and n 0 = i 0 + 1.

[0083] It should be explained that μ n1 , μ n2 can be obtained through the following steps: Sum and analyze the change rates of the comprehensive agricultural health flow index of the nth and (n + 1)th groups to obtain the sum value of the comprehensive agricultural health flow change of the nth group. Then, perform a proportion analysis on the change rates of the comprehensive agricultural health flow index of the nth and (n + 1)th groups respectively with the sum value of the comprehensive agricultural health flow change of the nth group. The result of the proportion analysis is the corresponding weighted coefficient.

[0084] can be obtained through the following steps: First, historical data. By analyzing the dynamic changes of the change rates of the comprehensive agricultural health flow index of adjacent two groups, use the statistical regression method to quantify the initial influence degree of the change rates of the comprehensive agricultural health flow index of adjacent two groups on the comprehensive agricultural health prediction index, so as to obtain the initial coefficient value. Then, based on the sensitivity analysis technology, adjust the value range of these coefficients in different environmental scenarios to ensure the applicability of the formula to diverse crop growth environments.

[0085] In this implementation plan, through the trend analysis and mean calculation of the change rate of the comprehensive agricultural health flow index, it is possible to accurately predict the health status of crops in the next monitoring period in the set farmland, and then help farmers make adjustments in advance, such as optimizing irrigation, fertilization and other measures. And by predicting the health changes of crops, farmers can avoid sudden problems, so as to ensure the continuity and stability of farmland management, and then improve production efficiency and crop yield. Secondly, through the weighted analysis of the change rates of the comprehensive agricultural health flow index in different monitoring periods, the adaptability of the farmland management strategy can be dynamically adjusted according to different stages in the crop growth process. Finally, through the weighted coefficient and the adjustment coefficient, it is possible to ensure that the prediction index is more in line with the actual situation, and then improve the accuracy of farmland management, and then achieve more precise agricultural management.

[0086] Please refer to Figure 3, an embodiment of the present invention provides a technical solution: a real-time monitoring system for crop growth, including: a data acquisition module, an image analysis module, a data analysis module, a prediction analysis module, and a growth warning module; the data acquisition module is used to continuously and real-time acquire the time series of crop image data and the time series data of soil status in a set farmland; the image analysis module is used to perform image analysis on the time series of crop image data in the set farmland to obtain the initial agricultural health flow index for each monitoring period in the set farmland; the data analysis module is used to perform data analysis on the time series of crop image data in the set farmland to obtain the soil adjustment intelligence index for each monitoring period in the set farmland, and perform comprehensive analysis in combination with the initial agricultural health flow index to obtain the comprehensive agricultural health flow index for each monitoring period in the set farmland; the prediction analysis module is used to perform prediction analysis on the comprehensive agricultural health flow index for each monitoring period in the set farmland to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland; the growth warning module is used to compare and analyze the comprehensive agricultural health flow index for the next monitoring period in the set farmland with a preset comprehensive agricultural health flow index threshold, and take corresponding warning measures based on the comparison and analysis results.

[0087] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A method for real-time monitoring of crop growth, characterized in that: The following steps are involved: Continuously and in real time, obtain the crop image data time series and soil status time series data in the set farmland, and perform data analysis respectively to obtain the initial agricultural health flow index and soil adjustment index of each monitoring period in the set farmland, and perform comprehensive analysis to obtain the comprehensive agricultural health flow index of each monitoring period in the set farmland; Predict and analyze the comprehensive agricultural health flow index of each monitoring period in the set farmland, obtain the comprehensive agricultural health flow index of the next monitoring period in the set farmland, and compare and analyze it with the preset comprehensive agricultural health flow index range, and take corresponding early warning measures based on the comparison and analysis results; The specific formula for calculating the comprehensive agricultural health flow index for each monitoring period in the set farmland is as follows: ； Among them, ZhW i To set the comprehensive agricultural health flow index of the i-th monitoring period in the farmland, CzH i TrX i They are the initial agricultural health flow index and soil intelligence index of the i-th monitoring period in the set farmland, α1, β1, α2, β2, β3 are the initial agricultural health coefficient, initial agricultural health adjustment coefficient, soil intelligence index, soil intelligence coefficient, soil intelligence adjustment coefficient, and interaction coefficient stored in the database, α1+α2=1, i=1, 2, 3,…, i0, i0 is the number of monitoring periods.

2. The method for real-time monitoring of crop growth according to claim 1, characterized in that: The crop image data time series includes the pixel value and two-dimensional coordinates of each pixel point in each monitoring period, and the soil state time series data includes the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value in each monitoring period.

3. The method for real-time monitoring of crop growth according to claim 2, characterized in that: The specific steps to obtain the initial agricultural health flow index for each monitoring period in the set farmland are as follows: Identify and analyze the pixel value and two-dimensional coordinates of each pixel point in each monitoring period in the set farmland to obtain a number of leaf areas and leaf stem areas in each monitoring period in the set farmland; Comprehensively analyzing the pixel value of each pixel point in each leaf area in each monitoring period in the set farmland, and obtaining the leaf green saturation index in each monitoring period in the set farmland; Read the two-dimensional coordinates of each leaf edge pixel point in each leaf area of ​​each monitoring period in the set farmland, and conduct a comprehensive analysis to obtain the leaf structure diversity index of each monitoring period in the set farmland; Read the pixel value and two-dimensional coordinates of each leaf stem pixel point in each leaf stem area in each monitoring period in the set farmland, and perform comprehensive analysis to obtain the leaf stem curvature index in each monitoring period in the set farmland; A comprehensive analysis was conducted on the leaf green saturation index, leaf structure diversity index and leaf stem curvature index of each monitoring period in the set farmland to obtain the initial agricultural health flow index of each monitoring period in the set farmland.

4. The method for real-time monitoring of crop growth according to claim 3, characterized in that: The specific steps for obtaining several leaf areas and leaf stem areas in each monitoring period in the set farmland are as follows: Performing color space conversion processing on the pixel value of each pixel point in each monitoring period in the set farmland to obtain the hue value, saturation value, and brightness value of each pixel point in each monitoring period in the set farmland, and performing threshold segmentation processing to obtain a number of color predicted leaf areas and color predicted leaf stem areas in each monitoring period in the set farmland; And grayscale processing is performed on the pixel value of each pixel point in each monitoring period in the set farmland; Perform edge detection processing based on the pixel value of each pixel point in each monitoring period in the set farmland after grayscale processing, and obtain a number of edge predicted leaf areas and edge predicted leaf stem areas in each monitoring period in the set farmland; The two-dimensional coordinates of each color predicted leaf area, color predicted leaf stem area, edge predicted leaf area, and edge predicted leaf stem area in each monitoring period in the set farmland are read respectively for comprehensive analysis to obtain several leaf areas and leaf stem areas in each monitoring period in the set farmland.

5. The method for real-time monitoring of crop growth according to claim 3, characterized in that: The specific formula for calculating the initial agricultural flow index for each monitoring period in the set farmland is as follows: ； Among them, CzH i YbH is the initial agricultural flow index for the i-th monitoring period in the farmland. i 、YdX i ,QwZ i They are the leaf green saturation index, leaf structure diversity index and leaf stem curvature index of the i-th monitoring period in the set farmland, Φ1, Φ2, Φ3, Φ4 and Φ5 are the green saturation adjustment coefficient, structure diversity adjustment coefficient, difference adjustment coefficient, curvature adjustment coefficient and ratio adjustment coefficient stored in the database, i=1, 2, 3,…, i0, where i0 is the number of monitoring periods.

6. The method for real-time monitoring of crop growth according to claim 3, characterized in that: The specific steps for obtaining the leaf stem curvature index of each leaf stem area in each monitoring period in the set farmland are as follows: Comprehensively analyzing the pixel value of each leaf stem pixel point in each leaf stem area in each monitoring period in the set farmland, to obtain the leaf stem texture complexity index of each leaf stem area in each monitoring period in the set farmland; Read the two-dimensional coordinates of each leaf stem edge pixel point of each leaf stem area in each monitoring period in the set farmland, and perform comprehensive analysis to obtain the leaf stem morphological tortuosity index of each leaf stem area in each monitoring period in the set farmland; The leaf stem texture complexity index and leaf stem morphology tortuosity index of each leaf stem area in each monitoring period in the set farmland are comprehensively analyzed to obtain the leaf stem tortuosity index of each monitoring period in the set farmland.

7. The method for real-time monitoring of crop growth according to claim 2, characterized in that: The specific steps to obtain the soil conditioning index for each monitoring period in the set farmland are as follows: Obtaining a soil state reference set in a set farmland, the soil state reference set including a soil suitability reference index, a soil compaction reference value, a soil humic acid content reference value, a soluble nutrient reference index, and a soil microbial activity reference value; A comprehensive analysis was conducted on the soil state reference set in the set farmland and the soil suitability index, soil compaction value, soil humic acid content value, soluble nutrient index, and soil microbial activity value of each monitoring period to obtain the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland; A comprehensive analysis was conducted on the soil suitability difference index, soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index, and soil microbial activity difference index of each monitoring period in the set farmland to obtain the soil intelligence index of each monitoring period in the set farmland.

8. The method for real-time monitoring of crop growth according to claim 7, characterized in that: The specific formula for calculating the soil suitability difference index and soil adjustment index for each monitoring period in the set farmland is as follows: ； Among them, TsC i To set the soil suitability difference index of the i-th monitoring period in the farmland, TsY i is the soil suitability index of the i-th monitoring period in the farmland, CsY is the soil suitability reference index in the farmland, TrX i To set the soil conditioning index of the i-th monitoring period in the farmland, KyC i , DhC i ,XkC i , YwC i They are the soil compaction difference index, soil humic acid content difference index, soluble nutrient difference index and soil microbial activity difference index of the i-th monitoring period in the set farmland, respectively; η1, η2, η3, η4, η5 and η6 are the soil suitability adjustment coefficient, soil compaction adjustment coefficient, humic acid adjustment coefficient, humic acid adjustment coefficient, microbial activity adjustment coefficient and soil interaction coefficient stored in the database, respectively; i=1,2,3,…,i0, where i0 is the number of monitoring periods.

9. The method for real-time monitoring of crop growth according to claim 1, characterized in that: The specific steps to obtain the comprehensive agricultural health flow index for the next monitoring period in the set farmland are as follows: Conduct trend analysis on the comprehensive agricultural health flow index of each monitoring period in the set farmland, and obtain the change rate of several groups of comprehensive agricultural health flow index in the set farmland; A mean analysis is performed on the change rates of several groups of comprehensive agricultural health flow indices in the set farmland to obtain the mean of the change rates of the comprehensive agricultural health flow indices in the set farmland, and a comprehensive analysis is performed to obtain the comprehensive agricultural health flow prediction index in the set farmland, which is regarded as the comprehensive agricultural health flow prediction index for the next monitoring period in the set farmland.

10. A crop growth real-time monitoring system, using the crop growth real-time monitoring method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, image analysis module, data analysis module, prediction analysis module, growth warning module; The data acquisition module is used to continuously and in real time acquire the time series of crop image data and soil state data in a set farmland; The image analysis module is used to perform image analysis on the time series of crop image data in the set farmland to obtain the initial crop health flow index of each monitoring period in the set farmland; The data analysis module is used to perform data analysis on the crop image data time series in the set farmland to obtain the soil adjustment index for each monitoring period in the set farmland, and to perform comprehensive analysis in combination with the initial agricultural health flow index to obtain the comprehensive agricultural health flow index for each monitoring period in the set farmland; The prediction and analysis module is used to predict and analyze the comprehensive agricultural health flow index of each monitoring period in the set farmland to obtain the comprehensive agricultural health flow index of the next monitoring period in the set farmland; The growth warning module is used to compare and analyze the comprehensive agricultural health flow index of the next monitoring period in the set farmland with the preset comprehensive agricultural health flow index threshold, and take corresponding warning measures based on the comparison and analysis results.

Images