Intelligent agricultural planting method and system

Abstract

The present application relates to the technical field of agricultural planting management, and discloses an intelligent agricultural planting method and system. The method includes: analyzing nutritional needs based on environmental sensing and plant image data to generate nutritional rationing data; predicting growth and formulating cultivation standards through seedling image features; extracting spectral parameters based on environmental condition analysis and canopy images to form an environmental control plan; calculating health indicators based on growth status assessment and pest and disease characteristics; optimizing management strategies through growth parameter time series analysis; and determining harvest timing and grade assessment based on plant maturity and quality characteristics. The present application organically combines environmental monitoring, growth diagnosis, quality assessment, and market analysis to provide more accurate and intelligent agricultural planting decision support.

Description

Technical Field

[0001] The present application relates to the field of agricultural planting management, and in particular to an intelligent agricultural planting method and system. Background Art

[0002] With the advancement of agricultural modernization, smart agricultural planting technology has been widely used. Existing smart agricultural planting technology mainly collects environmental data such as temperature, humidity, and light through environmental sensors, uses image recognition technology to monitor crop growth status, and performs simple threshold control based on these data. Some systems have also introduced machine learning algorithms to perform preliminary intelligent regulation of agronomic operations such as environmental control and irrigation and fertilization. At the same time, some systems have begun to try to incorporate market data and quality assessment into the management decision-making process to provide data support for agricultural production.

[0003] However, existing technologies have obvious shortcomings: although they can collect various types of agricultural production data, the way they process data is too simple, often based on fixed thresholds, and lack in-depth analysis of the correlation between data. Especially when faced with complex agricultural production scenarios, it is impossible to effectively integrate environmental data, growth data, quality data, and market data, resulting in a lack of systematic and forward-looking management decisions. Existing technologies also fail to make full use of historical planting data and cannot achieve effective inheritance and optimization of experience. Summary of the invention

[0004] The present application provides a smart agricultural planting method and system for improving the efficiency and accuracy of smart agricultural planting.

[0005] In the first aspect, the present application provides an intelligent agricultural planting method, which includes: analyzing and processing nutritional needs according to environmental sensor data and plant image data to obtain nutritional ratio parameter data, extracting and calculating features of nutrient formula through the nutritional ratio parameter data, and generating nutritional rationing data; performing multidimensional analysis of growth indicators based on the nutritional rationing data and seedling image feature data to obtain growth prediction data, calculating cultivation condition parameters based on the growth prediction data and leaf morphology feature data, and generating seedling standardized data; performing numerical analysis of environmental conditions according to the nutritional rationing data and the growth prediction data to obtain environmental control parameter data, and performing spectral parameter analysis through the environmental control parameter data and plant canopy image data. Perform feature extraction and calculation to form environmental regulation data; dynamically analyze the growth status according to the environmental control parameter data and the plant growth image sequence data to obtain growth assessment data, calculate the health index according to the growth assessment data and the pest and disease image feature data, and generate health diagnosis data; perform time series analysis on the growth parameters based on the growth assessment data and the health diagnosis data to obtain strategy optimization data, predict the periodic index through the strategy optimization data and the plant morphological evolution data to form management guidance data; perform feature analysis on the quality index according to the management guidance data and the plant maturity image data to obtain harvest timing data, calculate the grade parameter according to the harvest timing data and the crop quality feature data to generate quality assessment data.

[0006] In a second aspect, the present application provides an intelligent agricultural planting system, the intelligent agricultural planting system comprising:

[0007] The acquisition module is used to analyze and process the nutritional requirements according to the environmental sensor data and the plant image data to obtain the nutritional ratio parameter data, extract and calculate the characteristics of the nutrient formula through the nutritional ratio parameter data, and generate the nutritional ration data;

[0008] An analysis module, for performing multi-dimensional analysis on growth indicators based on the nutrient ration data and seedling image feature data to obtain growth prediction data, calculating cultivation condition parameters based on the growth prediction data and leaf morphology feature data to generate seedling standardized data;

[0009] An extraction module is used to perform numerical analysis on environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and to perform feature extraction and calculation on spectral parameters through the environmental control parameter data and plant canopy image data to form environmental regulation data;

[0010] A calculation module, used to dynamically analyze the growth status according to the environmental control parameter data and the plant growth image sequence data to obtain growth evaluation data, calculate the health index according to the growth evaluation data and the pest and disease image feature data, and generate health diagnosis data;

[0011] A prediction module, used to perform time series analysis on growth parameters based on the growth assessment data and the health diagnosis data to obtain strategy optimization data, and to predict periodic indicators through the strategy optimization data and plant morphological evolution data to form management guidance data;

[0012] The evaluation module is used to perform feature analysis on quality indicators based on the management guidance data and the plant maturity image data to obtain harvest timing data, calculate grade parameters based on the harvest timing data and the crop quality feature data, and generate quality evaluation data.

[0013] In the technical solution provided by this application, by establishing an association analysis mechanism between environmental sensor data and plant image data, an accurate assessment of nutritional needs is achieved, and nutritional rationing is made more accurate through feature extraction of nutritional ratio parameter data; the accuracy of growth status assessment is improved by multidimensional analysis of seedling image feature data and growth prediction data, and the standardization level of seedlings is significantly improved by the introduction of leaf morphological feature data; numerical analysis of environmental conditions is carried out in combination with nutritional rationing data and growth prediction data, and spectral feature extraction of plant canopy image data is used to achieve refined management of environmental regulation; a complete health diagnosis system is constructed through dynamic analysis of plant growth image sequence data and computational processing of pest and disease image feature data; a scientific management guidance system is established based on the time series analysis of growth assessment data and health diagnosis data, combined with plant morphological evolution data; and a comprehensive analysis of plant maturity image data and crop quality feature data achieves accurate judgment of harvest timing and objective quantification of quality assessment. The entire solution establishes a complete data processing chain from plant growth monitoring to management decision-making through deep fusion and analysis of multidimensional data, realizes precise management and intelligent decision-making of the agricultural production process, and improves the scientificity and efficiency of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0015] Figure 1 This is a schematic diagram of an embodiment of the intelligent agricultural planting method in the embodiment of the present application;

[0016] Figure 2 This is a schematic diagram of a standard growth curve in an embodiment of the present application;

[0017] Figure 3 This is a schematic diagram of an embodiment of the intelligent agricultural planting system in the embodiments of the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application provide a method and system for planting smart agriculture. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the intelligent agricultural planting method in the embodiment of the present application includes:

[0020] Step S101, analyzing and processing the nutritional requirements according to the environmental sensor data and the plant image data to obtain nutritional ratio parameter data, extracting and calculating the characteristics of the nutrient formula through the nutritional ratio parameter data, and generating nutritional ration data;

[0021] Step S102: Perform multidimensional analysis on growth indicators based on nutrient ration data and seedling image feature data to obtain growth prediction data, calculate cultivation condition parameters based on the growth prediction data and leaf morphology feature data, and generate seedling standardized data;

[0022] Step S103, numerically analyzing the environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and extracting and calculating the features of the spectral parameters through the environmental control parameter data and the plant canopy image data to form environmental regulation data;

[0023] Step S104: dynamically analyzing the growth status according to the environmental control parameter data and the plant growth image sequence data to obtain growth evaluation data, and calculating the health index according to the growth evaluation data and the pest and disease image feature data to generate health diagnosis data;

[0024] Step S105: Perform time series analysis on growth parameters based on growth assessment data and health diagnosis data to obtain strategy optimization data, and predict periodic indicators through strategy optimization data and plant morphological evolution data to form management guidance data;

[0025] Step S106: Perform feature analysis on quality indicators based on the management guidance data and the plant maturity image data to obtain harvest timing data, calculate grade parameters based on the harvest timing data and the crop quality feature data to generate quality assessment data.

[0026] It is understandable that the execution subject of the present application may be a smart agricultural planting system, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0027] Specifically, the temperature, humidity, and light data are collected through environmental sensors, and multi-angle images of plants are collected at the same time. The environmental sensor data is standardized, outliers are removed, and plant images are segmented to extract color, shape, and texture features to establish a plant growth status database. These data are analyzed for association, and the nutritional deficiency symptoms of plants are identified through feature extraction. For example, yellowing of leaves indicates nitrogen deficiency, and scorching of leaf edges indicates potassium deficiency. These feature information is matched with the nutritional demand database to calculate the ratio parameters of nitrogen, phosphorus, potassium and other nutrient elements required for the current growth stage. After feature extraction, the environmental sensor data and plant image data enter the data fusion stage, and weight calculations are performed on various types of data to generate nutritional ratio parameter data. Based on this, the nutrient prescription is refined and calculated, taking into account the differences in nutritional requirements in different growth periods, and forming nutritional rationing data.

[0028] The nutrient rationing data is correlated with the collected seedling images. The seedling images contain morphological features of key parts such as stems and leaves. Through image segmentation and feature extraction, various indicators characterizing the growth status are calculated. These indicators are compared with historical growth data to predict future growth trends and generate growth prediction data. Combined with leaf morphological feature data, including leaf area, number of leaves, leaf color and other parameters, the cultivation environment conditions are optimized and calculated to obtain the optimal values ​​of environmental parameters such as temperature, humidity, and light, forming seedling standardized data. Based on nutrient rationing data and growth prediction data, the impact of environmental factors on plant growth is comprehensively analyzed, and the optimal environmental control parameters are calculated through numerical simulation. Plant canopy images are collected, plant height, branch angle, leaf arrangement and other features are extracted, light energy utilization efficiency is calculated, spectral parameters are optimized, and environmental regulation data are generated.

[0029] Environmental control parameter data is combined with continuously collected plant growth image sequences to analyze dynamic changes in plant growth and development. Image processing technology is used to identify the growth status of leaves, stems, fruits and other organs, and to generate growth assessment data. At the same time, feature extraction is performed on pest and disease images to identify abnormal phenomena such as disease spots and pests, calculate health indicators, and form health diagnosis data.

[0030] Combined with growth assessment data and health diagnosis data, the plant growth parameters are analyzed in time series to identify growth trends and potential problems and form strategy optimization data. Through the continuous collection of plant morphology images, the changes in plant morphology over time are recorded, the growth cycle indicators are predicted, the corresponding management strategies are formulated, and management guidance data are generated. When the crops are close to maturity, the plant growth status is comprehensively evaluated through management guidance data, crop maturity images are collected, the color, size, shape and other characteristics of the fruits and seeds are analyzed, and the best time to harvest is determined. Combined with the quality characteristic data of crops, including appearance, nutritional content, taste and other indicators, the product grades are classified and evaluated to generate quality assessment data.

[0031] For example, in tomato planting, by analyzing the changes in the RGB values ​​of leaf color, when the G value of the leaf image decreases by 20%, it indicates nitrogen deficiency. Combined with environmental data such as temperature and humidity, the nitrogen fertilizer demand is calculated. Continuously take photos of tomato plants, extract growth parameters such as plant height and leaf area through image processing, establish a growth curve, and predict yield. Collect canopy images to calculate the leaf area index and optimize the lighting strategy. Identify early diseases through the characteristics of leaf spots and adjust management measures in time. Record quality indicators such as fruit color and hardness, determine the best harvest period, and form a digital planting plan.

[0032] In the embodiment of the present application, by establishing an association analysis mechanism between environmental sensor data and plant image data, an accurate assessment of nutritional needs is achieved, and nutritional rationing is made more accurate by extracting the characteristics of nutritional ratio parameter data; the accuracy of growth status assessment is improved by using multidimensional analysis of seedling image feature data and growth prediction data, and the standardization level of seedlings is significantly improved by introducing leaf morphological feature data; numerical analysis of environmental conditions is performed in combination with nutritional rationing data and growth prediction data, and spectral feature extraction of plant canopy image data is used to achieve refined management of environmental regulation; a complete health diagnosis system is constructed by dynamic analysis of plant growth image sequence data and computational processing of pest and disease image feature data; a scientific management guidance system is established based on the time series analysis of growth assessment data and health diagnosis data, combined with plant morphological evolution data; and a comprehensive analysis of plant maturity image data and crop quality feature data achieves accurate judgment of harvest timing and objective quantification of quality assessment. The entire scheme establishes a complete data processing chain from plant growth monitoring to management decision-making through deep fusion and analysis of multidimensional data, realizes precise management and intelligent decision-making of the agricultural production process, and improves the scientificity and efficiency of agricultural production.

[0033] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0034] (1) Collect images of plant leaves, stems, and roots, segment the images to generate organ feature data, match the organ feature data with environmental sensor data such as light, temperature, and humidity, and obtain plant physiological parameters;

[0035] (2) Correlation analysis is performed between plant physiological parameters and standard growth curve data to extract plant nutrient deficiency characteristics, quantify the plant nutrient deficiency characteristics, and construct a nutrient requirement index;

[0036] (3) Compare and analyze the nutrient demand index with historical nutrient response data, output nutrient ratio parameter data, perform cluster analysis on the nutrient ratio parameter data, and determine the nutrient characteristic range;

[0037] (4) Perform weighted calculation on the nutrient characteristic interval data, establish a nutrient formula matrix, extract features from the nutrient formula matrix, and identify the proportions of key nutrient elements;

[0038] (5) Cross-validate the ratio of key nutrients with growth stage data, construct nutritional balance indicators, perform numerical optimization on nutritional balance indicators, and form a nutrient allocation plan;

[0039] (6) Correct the calculation of the nutrient allocation plan and the real-time environmental data, adjust the nutrient allocation ratio coefficient, perform data verification based on the nutrient allocation ratio coefficient, and generate nutrient allocation data.

[0040] Specifically, a high-resolution camera is used to photograph various organs of the plant, including leaves, stems and roots. The collected images are preprocessed, and the various organ parts are separated and extracted through image segmentation technology: leaf area and chlorophyll content indicators are extracted from leaf images; stem diameter and internode distance data are extracted from stem images; main root length and branch number characteristics are extracted from root images to form an organ feature data set. At the same time, environmental parameters such as light intensity, temperature, and relative humidity are collected through environmental sensors to establish an environmental parameter data set. The organ feature data and the environmental parameter data are time-series matched, the organ growth response values ​​under different environmental conditions are calculated, and a plant physiological parameter matrix is ​​generated. After obtaining the plant physiological parameters, they are compared with the healthy growth data in the crop standard growth curve database. The standard growth curve data contains the standard physiological indicator values ​​of crops in each growth period under normal growth conditions. Such as Figure 2 As shown, it is a schematic diagram of the standard growth curve provided in the embodiment of the present application, which shows the dynamic change trend of three key parameters in the growth and development process of plants from the seedling stage to the fruiting stage. The solid line represents the plant height (cm), which shows a significant growth trend throughout the growth cycle, from about 10 cm in the seedling stage to 120 cm in the fruiting stage; the dotted line represents the leaf area (square centimeters), and its change trend is similar to that of the plant height, but the growth rate is relatively small, from about 5 square centimeters in the seedling stage to 90 square centimeters in the fruiting stage; the dotted line represents the chlorophyll content (mg / g), which remains relatively stable throughout the growth process, only slowly rising from 1 mg / g in the seedling stage to 7 mg / g in the fruiting stage. The figure clearly reflects that the plants undergo the most significant morphological changes during the vegetative growth period and the flowering period, and the growth rate tends to be flat in the fruiting period. Through data comparison, plant growth deviations are identified and nutrient deficiency characteristics are extracted. The following formula is used to quantitatively calculate the nutrient deficiency characteristics:

[0041]

[0042] in, is the nutritional requirement index, is the influencing factor of the ith physiological index, is the actual measured value of the ith physiological parameter, is the corresponding parameter value in the standard growth curve, is the weight coefficient, and n is the total number of physiological indicators.

[0043] Compare and analyze the calculated nutrient demand index with the nutrient response data in the historical database. The historical nutrient response data records the growth response of plants under different nutrient ratio conditions. Through data association analysis, the nutrient ratio parameter data required for the current growth stage is output. Cluster calculation is performed on these parameter data, similar nutrient demand characteristics are classified, and the nutrient characteristic range is determined.

[0044] The nutrient characteristic interval data were weighted and the nutrient formula matrix was established using the following formula:

[0045]

[0046] in, is the nutrient formula matrix element, is the absorption coefficient of the i-th nutrient element, is the demand coefficient of the jth growth stage, is the historical response intensity, is the current demand, is the historical average, is the environmental correction coefficient. Extract eigenvalues ​​from the nutrient formula matrix, identify the ratio relationship between different nutrients, and obtain the ratio of key nutrients. Cross-validate these ratio data with crop growth stage data to build an indicator system that reflects the balance of nutrient requirements in different growth periods. Use numerical optimization methods to adjust the ratio of each nutrient element and form a nutrient allocation plan.

[0047] The nutrient allocation scheme is analyzed in association with the real-time collected environmental data, and the nutrient allocation ratio is dynamically adjusted according to the current environmental conditions. The adjusted nutrient ratio coefficient is obtained through correction calculation, and the ratio coefficient is verified by data to ensure that it conforms to the physiological needs of plants and generate nutrient allocation data.

[0048] For example, in hydroponic tomato cultivation, yellowing of leaf morphology was found through image acquisition. Leaf color characteristics were extracted through image analysis. Comparison with standard leaf color data revealed that chlorophyll content was lower than the standard value. Plant physiological parameters were calculated by combining stem and root image analysis results and environmental monitoring data. These parameters were compared with standard growth curves to find nitrogen deficiency characteristics. The current nitrogen requirement was determined by calculating the nutrient requirement index. Combined with historical nutrient response data, the optimal nitrogen, phosphorus and potassium ratio was calculated. Taking into account the characteristics of the current growth period, the nutrient formula was optimized and corrected according to real-time light intensity, temperature and other environmental conditions to generate nutrient rationing data suitable for the current growth stage.

[0049] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0050] (1) The nutrient ration data and growth stage characteristics are integrated to collect data on seedling stem thickness, internode spacing, and leaf size to form seedling image feature data. The seedling image feature data is numerically counted to obtain growth benchmark parameters.

[0051] (2) Conduct correlation analysis on growth benchmark parameters and historical seedling data, extract key growth characteristics, associate key growth characteristics with environmental parameters, construct a growth indicator matrix, and obtain growth prediction data through multidimensional data processing;

[0052] (3) Collect seedling leaf images, extract data on leaf color, area, and shape, generate leaf morphological feature data, match the leaf morphological feature data with growth prediction data, and obtain growth status parameters;

[0053] (4) Divide the growth state parameters into intervals, determine the cultivation condition thresholds, conduct correlation analysis between the cultivation condition thresholds and environmental historical data to form cultivation condition parameters, and perform data verification on the cultivation condition parameters;

[0054] (5) Perform time series processing on the cultivation condition parameters and growth cycle data, calculate the seedling development index, perform data correction on the seedling development index, and form a standardized index;

[0055] (6) Screen the data of standardized indicators, establish a seedling quality evaluation system, compare the seedling quality evaluation system with the target parameters, and generate seedling standardized data.

[0056] Specifically, the nutrient rationing data is integrated into the growth stage feature analysis process. The nutrient rationing data contains the information on the ratio of various nutrient elements required for different growth periods, and the growth stage characteristics reflect the growth and development status of plants at different stages. Through data fusion, the corresponding relationship between nutrient supply and growth and development is established. At the same time, the morphological characteristics of seedlings are collected: the stem cross section is photographed with a high-definition camera to calculate the stem thickness data; the distance between two adjacent nodes is measured to obtain the internode data; the leaf image is collected and the number and size of leaves are extracted. These data together constitute the seedling image feature data set. Numerical statistical analysis is performed on these feature data, including mean calculation, standard deviation analysis, and coefficient of variation statistics, to form benchmark parameters reflecting the growth status of seedlings. The growth benchmark parameters are compared and analyzed with the historical seedling database, and the key indicators that have a significant impact on growth and development are identified through correlation calculation. These historical seedling data come from the records of healthy seedlings that have been successfully cultivated in the past, and include growth process data. Through data mining, key features that affect seedling growth, such as stem thickness growth rate and leaf expansion speed, are extracted. These key growth characteristics are associated with real-time environmental parameters such as temperature, light, and humidity to construct a multidimensional growth indicator matrix. The matrix is ​​processed with multidimensional data, and time series analysis methods are used to predict the growth trend of seedlings and generate growth prediction data.

[0057] Special image collection and analysis are performed on seedling leaves. Through image processing technology, the color RGB value of the leaves, the leaf projection area, the leaf contour characteristics and other data are extracted to generate a leaf morphological feature data set. These morphological feature data are matched and analyzed with the aforementioned growth prediction data to evaluate whether the current growth status meets expectations and obtain growth status parameters that reflect the overall growth of the seedlings. The growth status parameters are divided into numerical intervals, the normal range and critical values ​​of the growth indicators are established, and the thresholds of cultivation conditions such as temperature, humidity, and light are determined. These thresholds are correlated with environmental historical data to study the influence of different environmental conditions on seedling growth and form detailed cultivation condition parameters. The scientific nature and reliability of these parameters are ensured through data verification methods.

[0058] The cultivation condition parameters and growth cycle data are processed in time series, the development characteristics of seedlings at different growth stages are analyzed, and the development index reflecting the growth rate and quality of seedlings is calculated. The development index is corrected to eliminate the interference caused by environmental fluctuations and form a universal standardized indicator. Screening conditions are set for the standardized indicators to establish a seedling quality evaluation system that includes multiple dimensions such as growth rate, morphological characteristics, and resistance performance. The evaluation system is compared and analyzed with the pre-set target parameters to evaluate the quality of seedling cultivation and generate seedling standardized data.

[0059] For example, in the cultivation of cucumber seedlings, the nutrient ration data of cucumber seedlings is collected, including the ratios of nitrogen, phosphorus, potassium and other elements required at different growth stages. Growth parameters such as stem diameter and internode spacing are acquired through image acquisition, and combined with leaf feature analysis to generate a seedling feature data set. Comparing these data with historical success cases, it is found that the stem diameter growth rate and leaf area expansion rate are key growth indicators. Combined with greenhouse environmental data, a growth prediction model is constructed. Through leaf image analysis, leaf color and leaf area changes are monitored in real time to evaluate growth status. The optimal temperature and light range are determined based on historical data to form cultivation parameters. Dynamic monitoring of seedling growth is carried out, development indexes are calculated, and standardized data is generated according to target requirements to guide the precise regulation of the seedling cultivation process.

[0060] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0061] (1) Couple the nutrient ration data with the growth prediction data, extract the environmental influencing factors, perform numerical calculations on the environmental influencing factors, form an environmental condition matrix, split the environmental condition matrix, and obtain the environmental control parameter data;

[0062] (2) Conduct correlation analysis on moisture, temperature, and humidity data, establish a parameter response table, perform numerical verification on environmental control parameter data, form environmental benchmark data, perform time series processing on the environmental benchmark data, and obtain environmental change trends;

[0063] (3) Collect multi-angle images of plant canopies, extract canopy density data, associate the canopy density data with light intensity data, generate plant canopy image data, perform feature calculations on the plant canopy image data, and obtain photosynthetic parameters;

[0064] (4) Fusing photosynthetic parameters with environmental spectral data, extracting spectral characteristic values, performing numerical operations on the spectral characteristic values ​​to form spectral parameters, comparing and analyzing the spectral parameters with historical data, and establishing spectral regulation indicators;

[0065] (5) Data stratification of spectral control indicators, calculation of spectral compensation coefficients, matching of spectral compensation coefficients with environmental factors, forming control strategy data, and numerical verification of the control strategy data;

[0066] (6) Comprehensively process the control strategy data and environmental control parameter data, construct an environmental response matrix, optimize the environmental response matrix data, and generate environmental control data.

[0067] Specifically, the nutrient ration data and growth prediction data are coupled. The nutrient ration data contains the ratio information of various nutrient elements, and the growth prediction data contains the plant growth and development trend data. Through data coupling, the correlation between the two sets of data is analyzed, and environmental influencing factors such as temperature, humidity, and light are extracted. These environmental influencing factors are numerically calculated using the following formula:

[0068]

[0069] in, is the environmental condition matrix element, is the influence weight of the kth environmental factor, is the response intensity of the i-th nutrient element to the k-th environmental factor, is the sensitivity of the jth growth prediction index to the kth environmental factor, is the coefficient of variation of the environmental factor, is the data coupling degree, and n is the total number of environmental factors. Split the environmental condition matrix according to different environmental factors to obtain environmental control parameter data such as temperature, humidity, and light. Perform correlation analysis on the collected moisture, temperature, and humidity data to study the interaction between different environmental factors. Establish a response relationship table between various parameters and record the plant growth response data under different combinations of environmental conditions. Compare and verify the environmental control parameter data obtained in the previous step with the response relationship table to form environmental baseline data. Perform time series analysis on the environmental baseline data to identify the changing rules of environmental parameters and obtain environmental change trends.

[0070] The plant canopy is photographed in stereo using a multi-angle camera array, and canopy density data such as plant density and leaf distribution are extracted through image processing methods. The canopy density data is correlated with the real-time measured light intensity data to generate plant canopy image data that reflects the relationship between canopy structure and light distribution. Further feature calculations are performed on the plant canopy image data to obtain photosynthetic parameters such as chlorophyll content and light energy utilization.

[0071] The photosynthetic parameters and environmental spectral data were fused and the spectral features were extracted using the following formula:

[0072]

[0073] in, is the spectral characteristic value, is the weight coefficient of the lth band, is the performance of the mth photosynthetic parameter in the lth band, is the intensity of the nth ambient spectrum in the lth band, is the band reference value, is the fusion coefficient, and p is the number of spectral bands. The spectral characteristic values ​​are numerically calculated to form spectral parameters, which are compared and analyzed with historical spectral data to establish spectral control indicators.

[0074] The spectral control index is layered according to different spectral bands, and the spectral compensation coefficients between each layer are calculated. The spectral compensation coefficients are matched and analyzed with the environmental factor data such as temperature and humidity to form the control strategy data, and the accuracy of the strategy is ensured through numerical verification. The control strategy data and the environmental control parameter data are comprehensively processed to construct a response matrix containing various environmental factors. The environmental response matrix is ​​optimized to coordinate the balance relationship between various environmental parameters and generate environmental control data.

[0075] For example, in greenhouse tomato cultivation, the nutritional demand data of tomatoes is associated with the growth prediction data to analyze the degree of influence of temperature, humidity, and light on plant growth. Through the calculation of the environmental condition matrix, the control parameters that the daytime temperature should be controlled within a certain range are obtained. By analyzing the greenhouse environmental data, it is found that the temperature and humidity are negatively correlated, and an environmental parameter response table is established based on this. By photographing tomato plants from multiple angles, canopy parameters such as leaf area index are calculated, and the light energy utilization efficiency is evaluated in combination with light intensity data. The effects of different spectral bands on tomato growth are analyzed, and the composition of the supplementary light spectrum is calculated. According to the spectral characteristic values ​​and environmental conditions, a supplementary light strategy is formulated to form an environmental control plan.

[0076] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0077] (1) Time-series matching of environmental control parameter data and plant growth status data, extracting key growth time points, collecting images of plants at different growth stages, generating plant growth image sequence data, extracting features from the plant growth image sequence data, and obtaining growth change parameters;

[0078] (2) Compare and calculate the growth change parameters with the standard growth data, establish a growth deviation matrix, perform numerical analysis on the growth deviation matrix, form growth status indicators, perform data correlation on the growth status indicators, and obtain growth evaluation data;

[0079] (3) Collect images of disease spots on plant leaves and stems, extract disease spot feature points, match the disease spot feature points with standard disease data, generate disease and insect pest image feature data, classify the disease and insect pest image feature data, and obtain disease type parameters;

[0080] (4) Fuse the growth assessment data with the disease type parameters to construct a health assessment matrix, perform numerical operations on the health assessment matrix to form health indicators, and compare and analyze the health indicators with historical data;

[0081] (5) Stratify the data of health indicators, calculate the health risk coefficient, associate the health risk coefficient with environmental factors, form diagnostic strategy data, and perform numerical verification on the diagnostic strategy data;

[0082] (6) Comprehensively process the diagnostic strategy data and growth assessment data, construct a health response matrix, optimize the health response matrix data, and generate health diagnostic data.

[0083] Specifically, the environmental control parameter data and the plant growth status data are matched and aligned in the time dimension. By analyzing the correspondence between environmental parameters and plant growth status at different time points, the key time nodes of growth and development, such as the seedling stage, tillering stage, and heading stage, are extracted. For these key time points, a high-resolution camera is used to take regular and fixed-point photos of the plants, record the morphological characteristics of different growth stages, and generate continuous plant growth image sequence data. Feature extraction is performed on these image sequences, including changes in key growth indicators such as plant height, stem thickness, and leaf area, to obtain growth change parameters. The growth change parameters are compared and calculated with the benchmark data in the standard growth database, and the growth deviation matrix is ​​established using the following formula:

[0084]

[0085] in, is the growth deviation matrix element, is the weight factor at the tth time point, is the measured value of the ith growth parameter at time t, is the reference value of the jth standard parameter at time t, is the time node benchmark value, is the matching coefficient, and q is the total number of time nodes. The growth status index is obtained through numerical analysis, and the index data is correlated and analyzed to generate growth evaluation data.

[0086] When collecting plant disease images, focus on abnormal features on the surface of leaves and stems. Use image segmentation technology to extract feature points of the diseased area, including the shape, color, texture and other features of the diseased area. Match these diseased feature points with the standard features in the plant disease database to generate disease and insect pest image feature data. Classify and analyze the feature data, identify the specific disease type, and obtain the disease type parameters. Fuse the growth assessment data with the disease type parameters to construct an assessment matrix that reflects the overall health status of the plant. Perform numerical operations on the matrix to calculate the health index. Compare and analyze the health index with the historical cultivation data to evaluate the current health status of the plant.

[0087] The health indicators are stratified according to the degree of disease and risk level, and the health risk coefficient is calculated. The calculation formula of the risk coefficient is as follows:

[0088]

[0089] in, is the health risk factor, is the weight of the kth health level, is the value of the health indicator at the kth level, Environmental stress factors, is the health threshold, is the risk assessment factor, and r is the health level. The health risk coefficient is correlated with environmental factor data such as temperature and humidity to form diagnostic strategy data, and numerical verification is performed. The diagnostic strategy data and growth assessment data are processed comprehensively to construct a response matrix including growth status and health status. The matrix is ​​optimized to coordinate the relationship between various indicators and generate plant health diagnostic data.

[0090] For example, in greenhouse sweet pepper cultivation, environmental sensors are used to collect environmental data such as temperature and humidity in the greenhouse, and sweet pepper plants are photographed regularly. During key periods such as 10 days after transplanting, flowering period, and fruiting period, the growth of plants is recorded. The images are analyzed to obtain growth parameters such as plant height growth rate and leaf area development. These parameters are compared with the standard growth curve, and it is found that the growth rate of the plants is slow. When the plants are inspected for diseases, spots are found on the leaves. The shape and color characteristics of these spots are identified through image analysis, and they are confirmed to be bacterial spot disease after comparison with the disease database. Combining the slow growth and disease characteristics, a higher health risk coefficient is calculated. Analysis of relevant environmental data found that the humidity in the greenhouse was too high, and based on this, a control strategy to reduce humidity and strengthen ventilation was formulated to form a plant health management plan.

[0091] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0092] (1) Correlate the growth assessment data with the health diagnosis data in time series, extract the characteristic points of the growth stages, perform numerical statistics on the characteristic points of the growth stages to form a growth parameter matrix, perform time series analysis on the growth parameter matrix, and obtain growth fluctuation indicators;

[0093] (2) Compare and calculate the growth fluctuation index with the historical growth data, establish a parameter optimization matrix, perform numerical analysis on the parameter optimization matrix, form decision parameters, verify the decision parameters with data, and obtain strategy optimization data;

[0094] (3) Collect images of plant morphological changes during the growth period, extract morphological feature points, match the morphological feature points with the growth cycle data, generate plant morphological evolution data, classify the plant morphological evolution data, and obtain morphological type parameters;

[0095] (4) Fusing the strategy optimization data with the pattern type parameters, constructing a cycle evaluation matrix, performing numerical operations on the cycle evaluation matrix, forming cycle indicators, and comparing and analyzing the cycle indicators with historical data;

[0096] (5) Data stratification of periodic indicators, calculation of periodic adjustment coefficients, data association of periodic adjustment coefficients with environmental factors to form management strategy data, and numerical verification of management strategy data;

[0097] (6) Comprehensively process the management strategy data and strategy optimization data, construct a management response matrix, optimize the management response matrix data, and generate management guidance data.

[0098] Specifically, a time series correlation analysis is performed on the growth assessment data and health diagnosis data. The growth assessment data includes indicators such as plant growth rate and morphological development, and the health diagnosis data includes information such as pest and disease status and physiological abnormalities. Through data time series matching, key growth stage feature points are identified, such as feature data of seedling, tillering, heading and other periods. These feature points are statistically analyzed, and the growth parameters of each period are calculated to form a parameter matrix describing the plant growth and development process. The matrix is ​​analyzed in time series to study the dynamic change law of plant growth and development and obtain indicators reflecting growth fluctuations. The growth fluctuation indicators are compared and analyzed with historical cultivation data. The historical data contains growth records of the same variety under similar cultivation conditions. Through data comparison, a parameter optimization matrix is ​​established, which reflects the difference between the current growth status and the historical optimal growth state. The matrix is ​​numerically analyzed to extract the key parameters that need to be optimized and adjusted, and form decision parameters to guide cultivation management. The decision parameters are data verified to ensure that they conform to the physiological and ecological characteristics of plants and obtain strategy optimization data.

[0099] The image acquisition device is used to continuously monitor the entire growth period of the plant. The collected image information includes the dynamic changes of morphological characteristics such as plant height, branches, and leaves. Morphological feature points such as stem elongation rate, leaf expansion angle, and number of branches are extracted from these images. These morphological feature points are paired with standard growth cycle data for analysis to generate a data set describing the changes in plant morphology over time. These data are classified and analyzed to determine the morphological feature types at different growth stages and obtain morphological type parameters. The strategy optimization data and morphological type parameters are fused. The strategy optimization data reflects the direction in which growth management needs to be improved, while the morphological type parameters provide the actual performance of plant growth and development. Through data fusion, a matrix for evaluating the plant growth cycle is constructed. The matrix is ​​numerically calculated to obtain indicators reflecting the characteristics of the growth cycle. These indicators are compared with historical cultivation experience data to evaluate whether the current growth cycle is within a reasonable range.

[0100] The cycle indicators are stratified according to different growth periods, and the growth cycle adjustment coefficients of each period are calculated. These adjustment coefficients are correlated with environmental factors such as temperature, light, and moisture to study the influence of environmental conditions on the growth cycle and form detailed management strategy data. The management strategy data is numerically verified to ensure the feasibility of the strategy.

[0101] The management strategy data and strategy optimization data are integrated and processed to construct a comprehensive management response matrix. The matrix contains various growth regulation indicators and their interrelationships. Through the data optimization algorithm, the relationship between various management elements is balanced to generate data to guide cultivation management.

[0102] For example, in rice cultivation, the growth assessment data and health diagnosis data of rice are analyzed to identify key growth nodes such as after transplanting, tillering period, and heading period. The growth parameters such as plant height, number of tillers, and leaf area in each period are counted, and it is found that the growth rate in the tillering period fluctuates. These data are compared with the records of the same period in previous years to determine the management parameters that need to be optimized. Images of the rice growth process are collected, and morphological characteristics such as stem elongation and leaf angle are analyzed. Compared with the standard growth process, it is found that the time for the conversion from vegetative growth to reproductive growth is slightly delayed. Combined with the growth strategy and morphological characteristic analysis, the growth process adjustment coefficient is calculated. According to the current environmental conditions such as temperature and light, a water and fertilizer management strategy is formulated to form a cultivation management plan. The entire process realizes closed-loop control from data collection and analysis to management decision-making.

[0103] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0104] (1) Match the management guidance data with the crop growth stage data in time series, extract the feature points of maturity, collect images of plant fruits and seeds, generate plant maturity image data, extract features from the plant maturity image data, and obtain maturity parameters;

[0105] (2) Compare and calculate the maturity parameters with the quality standard data, establish a quality assessment matrix, perform numerical analysis on the quality assessment matrix, form quality indicators, and associate the quality indicators with data to obtain harvest timing data;

[0106] (3) Collect images of crop appearance, size, and color, extract quality feature points, match the quality feature points with standard quality data, generate crop quality feature data, classify the crop quality feature data, and obtain quality type parameters;

[0107] (4) Fusing the harvest timing data with the quality type parameters, constructing a grade evaluation matrix, performing numerical operations on the grade evaluation matrix to form grade parameters, and comparing and analyzing the grade parameters with historical data;

[0108] (5) Stratify the data of the grade parameters, calculate the comprehensive quality coefficient, associate the comprehensive quality coefficient with the market factors, form the evaluation strategy data, and perform numerical verification on the evaluation strategy data;

[0109] (6) Comprehensively process the assessment strategy data and the harvest timing data, construct a quality response matrix, optimize the quality response matrix, and generate quality assessment data.

[0110] Specifically, the management guidance data is matched with the crop growth stage data in time series. The management guidance data contains the cultivation management requirements for the entire growth period, while the growth stage data records the stage characteristics of plant growth and development. Through data matching analysis, key feature point data close to maturity are extracted, such as grain filling degree, fruit coloring degree and other indicators. A high-resolution camera is used to shoot plant fruits or seeds from multiple angles, collect feature images such as fruit size, color, and morphology, and generate a plant maturity image dataset. Feature extraction is performed on these image data, including features such as fruit size, color uniformity, and surface gloss, to obtain parameters reflecting the degree of maturity. The maturity parameters are compared and calculated with the crop quality standard database. The quality standard data contains reference values ​​for various quality indicators of agricultural products of different grades. Through comparative analysis, an evaluation matrix reflecting the current crop quality status is established. Numerical analysis is performed on the matrix to extract key quality indicators, such as water content, sugar content, and hardness. These quality indicators are correlated with historical harvest experience data to obtain judgment data for the best harvest time.

[0111] Use a multispectral camera to collect the overall image of crops, focusing on recording features such as appearance integrity, fruit or seed size, and surface color. Extract quality feature points from these images, including data such as the shape, size, and color distribution of agricultural products. Match and analyze these feature points with the standard quality database to generate a crop quality feature data set. Classify these feature data to distinguish the characteristics of different quality grades and obtain quality type parameters. Fusion the harvest timing data and quality type parameters. The harvest timing data reflects the best time for harvesting, and the quality type parameters reflect the quality characteristics of the product. Through data fusion, a matrix for evaluating product grades is constructed. Numerical calculations are performed on the matrix to obtain grade parameters for different quality indicators. These parameters are compared with historical high-quality agricultural product data to evaluate the current product quality level.

[0112] The grade parameters are layered according to different quality factors, and the coefficient that comprehensively reflects the product quality is calculated. The comprehensive quality coefficient is correlated with market supply and demand, price trend and other market factor data to study the matching relationship between quality and market demand and form product evaluation strategy data. The evaluation strategy data is numerically verified to ensure the rationality of the strategy.

[0113] The evaluation strategy data and harvest timing data are integrated and processed to construct a response matrix that reflects product quality. The matrix contains various indicators of product quality and their temporal changes. Through the data optimization algorithm, the relationship between various quality factors is balanced to generate quality evaluation data.

[0114] For example, by analyzing the management guidance data and growth stage data of strawberries, we can identify the characteristics of maturity, such as the fruit turning from green to red and the rapid accumulation of sugar content. We can collect images of strawberry fruits, analyze the characteristics of fruit size, redness, glossiness, etc., and obtain maturity parameters. By comparing these parameters with the standard quality data, we can find that the fruit is evenly colored and of moderate size. We can collect images of the appearance of strawberries, analyze the shape, size, color and other characteristics, and determine the quality grade by comparing with the standard quality requirements. Combined with the current market conditions, we can calculate the comprehensive quality score. According to the scoring results and the best harvesting time, we can formulate a harvesting plan and generate a quality assessment plan. In this way, we can achieve data management of the entire process from fruit maturity judgment to harvesting decision.

[0115] In a specific embodiment, the process of performing the step of stratifying the level parameters may specifically include the following steps:

[0116] (1) Sort the grade parameters numerically, establish a quality hierarchy matrix, divide the quality hierarchy matrix into intervals, generate quality stratified data, perform threshold processing on the quality stratified data, and obtain stratified boundary points;

[0117] (2) Conduct correlation analysis on the stratified boundary points and historical quality data, establish a quality evaluation matrix, calculate the characteristics of the quality evaluation matrix, form quality weight parameters, perform numerical verification on the quality weight parameters, and obtain the comprehensive quality coefficient;

[0118] (3) Collect market price, supply and demand, and seasonal data, extract key market features, match key market features with historical transaction data, generate market factor data, perform trend analysis on market factor data, and obtain market volatility parameters;

[0119] (4) Data coupling of the comprehensive quality coefficient and the market volatility parameter to construct a strategy calculation matrix, numerically analyze the strategy calculation matrix to form strategy indicators, and then perform data screening on the strategy indicators;

[0120] (5) Perform time series processing on the strategy indicators, calculate the strategy adjustment coefficient, fuse the strategy adjustment coefficient with the environmental factors to form evaluation strategy data, and compare the evaluation strategy data with the benchmark data;

[0121] (6) Calculate the reliability of the evaluation strategy data, construct a verification indicator matrix, optimize the verification indicator matrix, and confirm the evaluation strategy data.

[0122] Specifically, the grade parameters are numerically sorted. The grade parameters include the quality index data of the size, color, shape, etc. of the agricultural products. These parameters are sorted from high to low according to the quality requirements to establish a quality hierarchy matrix. The matrix is ​​divided into intervals according to the quality standards, and the characteristic data of different quality levels are grouped to generate quality stratified data. Threshold conditions are set for the stratified data to determine the boundaries between different quality grades and obtain the stratified boundary points. The obtained stratified boundary points are compared and analyzed with the historical quality evaluation data. The historical quality data includes the quality evaluation records of similar agricultural products in different production seasons. Through correlation calculation, an evaluation matrix reflecting the current product quality level is established. The eigenvalues ​​of the matrix are calculated to extract the weight coefficients of different quality indicators. These weight parameters are numerically verified to ensure their scientificity and reliability, and the comprehensive quality coefficient is obtained.

[0123] Through market monitoring, we collect information such as market price data, supply and demand relationship data, seasonal change data, etc. of current agricultural products. We extract key market characteristics from these data, such as price trends, supply and demand ratios, seasonal fluctuations, etc. We match and analyze these market characteristics with the historical transaction database to generate factor data that reflects market conditions. We conduct trend analysis on market factor data, study the laws of market changes, and obtain parameters that reflect market fluctuations.

[0124] The quality comprehensive coefficient and the market volatility parameter are coupled with data. The quality comprehensive coefficient reflects the quality level of the product, and the market volatility parameter reflects the changes in market demand. Through data coupling, a calculation matrix to guide sales strategy is constructed. Numerical analysis is performed on the matrix to extract key strategy indicators. Data screening is performed on these indicators to retain the most valuable indicators. Time series processing is performed on the selected strategy indicators, the time law of strategy execution is analyzed, and the strategy adjustment coefficient is calculated. The calculation process uses the following formula:

[0125]

[0126] in, Adjust data for strategy, is the weight coefficient of the mth time node, is the strategy indicator value at the mth time point, is the environmental factor value at the mth time point, is the reference value, is the optimization coefficient, and s is the total number of time nodes. The calculated strategy adjustment coefficient is integrated with environmental factor data such as temperature and humidity to form the evaluation strategy data, and compared with historical benchmark data for verification. The evaluation strategy data is subjected to reliability verification calculations to construct a matrix containing multiple verification indicators. The verification matrix is ​​subjected to data optimization processing to ensure the accuracy and feasibility of the evaluation strategy and confirm the evaluation strategy data.

[0127] For example, in grape cultivation, the quality grade parameters of grapes are analyzed, including fruit size, sugar content, color and other indicators, to establish quality grading standards. Through historical data analysis, the boundary values ​​of different quality grades are determined, such as the dividing points of super large fruit, medium fruit and small fruit. These grading data are compared with the historical high-quality fruit data to calculate the quality weight coefficient. At the same time, market data is collected to analyze the price change patterns in different periods and find the price difference between early-maturing grapes and late-maturing grapes. Combine quality and market data to formulate differentiated sales strategies. According to the current environmental conditions such as temperature and humidity, the harvesting and sales plans are adjusted to form an evaluation strategy plan. In this way, data management of the entire process from quality evaluation to market strategy formulation is achieved.

[0128] In a specific embodiment, the process of performing the time series processing step on the strategy indicator may specifically include the following steps:

[0129] (1) Conduct time series analysis on strategy indicators, extract trend feature points, perform statistical processing on trend feature points, generate time series change data, conduct periodic analysis on time series change data, and obtain fluctuation law parameters;

[0130] (2) Calculate the correlation between the volatility law parameters and the historical strategy data, establish a strategy response matrix, extract features from the strategy response matrix, form adjustment weight parameters, perform numerical calculations on the adjustment weight parameters, and obtain the strategy adjustment coefficients;

[0131] (3) Collect temperature, humidity, and light intensity data, extract environmental characteristic values, match the environmental characteristic values ​​with historical environmental data, generate environmental factor data, perform correlation analysis on the environmental factor data, and obtain environmental impact parameters;

[0132] (4) Integrate the data of the strategy adjustment coefficient and the environmental impact parameter, construct a fusion calculation matrix, perform numerical operations on the fusion calculation matrix, form a fusion index, and perform data verification on the fusion index;

[0133] (5) Optimize the fusion indicators, calculate the evaluation coefficients, match the evaluation coefficients with historical data, form evaluation strategy data, and perform numerical verification on the evaluation strategy data;

[0134] (6) Compare the evaluation strategy data with the planting benchmark data, construct a comparison matrix, perform numerical analysis on the comparison matrix, and generate the evaluation strategy data.

[0135] Specifically, the strategy indicator data is analyzed in time series. The strategy indicator includes the data of various management measures in the planting process, such as the amount of fertilizer, the amount of irrigation, the temperature control and other parameters. Through the time series analysis method, the key trend feature points are extracted, such as the peak point of fertilizer application and the irrigation cycle point. These trend feature points are statistically processed, and the statistical quantities such as the mean value and variance are calculated to generate time series data describing the change law of management measures. The time series change data is analyzed for periodicity, the repetition law of management measures is identified, and the parameters reflecting the fluctuation characteristics of the data are obtained. The fluctuation law parameters are analyzed for correlation with the historical planting strategy data. The historical strategy data records the management experience in the past planting process, including the management measures and their effects at different growth stages. Through correlation calculation, the response relationship matrix between the current management strategy and the historical experience is established. The matrix is ​​analyzed for characteristics, the management elements with significant influence are extracted, and the weight parameters for adjusting the management measures are formed. The weight parameters are numerically calculated and verified to obtain the coefficients guiding the strategy adjustment.

[0136] Use sensor arrays to monitor the planting environment in real time and collect environmental data such as temperature, relative humidity, and light intensity. Extract key values ​​that reflect environmental characteristics from these data, such as daily average temperature and accumulated light. Match and analyze environmental characteristic values ​​with historical environmental databases to generate environmental factor data sets. Perform correlation analysis on environmental factor data, study the interactive relationship between different environmental factors, and obtain parameters that reflect environmental impacts. Integrate the strategy adjustment coefficient and environmental impact parameters. The strategy adjustment coefficient reflects the optimization direction of management measures, while the environmental impact parameters reflect the restrictive role of environmental conditions. Construct a fusion calculation matrix through data integration. Perform numerical operations on the matrix to generate fusion indicators that comprehensively consider management and environmental factors. Perform data verification on fusion indicators to ensure their accuracy and reliability.

[0137] The fusion index is optimized and the evaluation coefficient is obtained through numerical calculation. The evaluation coefficient is matched with the historical management experience data to form an evaluation strategy data set. The evaluation strategy data is numerically verified to ensure the feasibility of the strategy.

[0138] Compare and analyze the evaluation strategy data with the planting benchmark data. The planting benchmark data contains standard indicators of crop growth and development. Through data comparison, a comparison matrix for evaluating the effectiveness of the strategy is constructed. Numerical analysis is performed on the matrix to generate evaluation strategy data to guide agricultural production.

[0139] For example, in greenhouse cultivation of tomatoes, management data during the growth period of tomatoes are collected, including fertilization records, temperature control data, etc. By analyzing the temporal variation patterns of these data, it is found that there are significant differences in the amount of fertilizer applied during the vegetative growth period and the reproductive growth period. The current management data is compared with the successful planting experience of previous seasons to determine the weights of management measures at different growth stages. Environmental data is collected through temperature and humidity sensors, and it is found that the daily temperature difference is closely related to fruit quality. Combining management strategies and environmental conditions, a temperature control plan is formulated. The plan is verified and optimized to form a planting management strategy. The entire process realizes closed-loop management from data collection and analysis to strategy formulation.

[0140] The above describes the smart agricultural planting method in the embodiment of the present application. The following describes the smart agricultural planting system in the embodiment of the present application. Figure 3 In the embodiment of the present application, an embodiment of the intelligent agricultural planting system includes:

[0141] The acquisition module 201 is used to analyze and process the nutritional requirements according to the environmental sensor data and the plant image data to obtain the nutritional ratio parameter data, extract and calculate the characteristics of the nutrient formula through the nutritional ratio parameter data, and generate the nutritional ration data;

[0142] An analysis module 202 is used to perform multi-dimensional analysis on growth indicators based on the nutrient ration data and seedling image feature data to obtain growth prediction data, calculate cultivation condition parameters based on the growth prediction data and leaf morphology feature data, and generate seedling standardized data;

[0143] An extraction module 203 is used to perform numerical analysis on environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and to perform feature extraction and calculation on spectral parameters through the environmental control parameter data and the plant canopy image data to form environmental regulation data;

[0144] The calculation module 204 is used to dynamically analyze the growth state according to the environmental control parameter data and the plant growth image sequence data to obtain growth evaluation data, calculate the health index according to the growth evaluation data and the pest image feature data, and generate health diagnosis data;

[0145] Prediction module 205, used to perform time series analysis on growth parameters based on the growth assessment data and the health diagnosis data, obtain strategy optimization data, and predict periodic indicators through the strategy optimization data and plant morphological evolution data to form management guidance data;

[0146] The evaluation module 206 is used to perform feature analysis on the quality index according to the management guidance data and the plant maturity image data to obtain harvest timing data, calculate the grade parameters according to the harvest timing data and the crop quality feature data, and generate quality evaluation data.

[0147] Through the synergy of the above components, by establishing the correlation analysis mechanism of environmental sensor data and plant image data, the accurate assessment of nutritional needs is achieved, and the nutritional ratio parameter data feature extraction is used to make the nutritional ratio more accurate; the multidimensional analysis of seedling image feature data and growth prediction data is used to improve the accuracy of growth status assessment, and the standardization level of seedlings is significantly improved by introducing leaf morphological feature data; the environmental conditions are numerically analyzed in combination with nutritional ratio data and growth prediction data, and the spectral feature extraction of plant canopy image data is used to achieve refined management of environmental regulation; a complete health diagnosis system is constructed through dynamic analysis of plant growth image sequence data and computational processing of pest and disease image feature data; a scientific management guidance system is established based on the time series analysis of growth assessment data and health diagnosis data, combined with plant morphological evolution data; and the comprehensive analysis of plant maturity image data and crop quality feature data achieves accurate judgment of harvest timing and objective quantification of quality assessment. Through the deep integration and analysis of multidimensional data, the entire solution establishes a complete data processing chain from plant growth monitoring to management decision-making, realizes precise management and intelligent decision-making of the agricultural production process, and improves the scientificity and efficiency of agricultural production.

[0148] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A smart agricultural planting method, characterized in that: The smart agricultural planting method comprises: The nutritional requirements are analyzed and processed according to the environmental sensor data and the plant image data to obtain the nutritional ratio parameter data, and the nutrient formula is feature extracted and calculated through the nutritional ratio parameter data to generate the nutritional distribution data, including: collecting plant leaves, stems, and root system images, segmenting the images to form organ feature data, matching the organ feature data with the environmental sensor data of light, temperature, and humidity to obtain plant physiological parameters; correlating the plant physiological parameters with the standard growth curve data to extract the plant nutritional deficiency characteristics, quantifying the plant nutritional deficiency characteristics, and constructing the nutritional requirement index; comparing the nutritional requirement index with the plant physiological parameters to obtain the nutritional deficiency characteristics of the plant; and Comparative analysis is performed on historical nutrition response data to output nutrition ratio parameter data, cluster analysis is performed on the nutrition ratio parameter data to determine nutrient characteristic intervals; weighted calculation is performed on the nutrient characteristic interval data to establish a nutrient formula matrix, feature extraction is performed on the nutrient formula matrix, and key nutrient element ratios are identified; cross-validation is performed on the key nutrient element ratios and growth stage data to construct nutrition balance indicators, and the nutrition balance indicators are numerically optimized to form nutrition allocation plans; correction calculation is performed on the nutrition allocation plans and real-time environmental data to adjust the nutrition ratio coefficients, data verification is performed based on the nutrition ratio coefficients, and nutrition rationing data is generated; Performing a multi-dimensional analysis on growth indicators based on the nutrient ration data and seedling image feature data to obtain growth prediction data, calculating cultivation condition parameters based on the growth prediction data and leaf morphology feature data to generate seedling standardized data; Performing numerical analysis on environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and performing feature extraction and calculation on spectral parameters through the environmental control parameter data and plant canopy image data to form environmental regulation data; Dynamically analyzing the growth status according to the environmental control parameter data and the plant growth image sequence data to obtain growth assessment data, and calculating the health index according to the growth assessment data and the pest and disease image feature data to generate health diagnosis data; Based on the growth assessment data and the health diagnosis data, a time series analysis of growth parameters is performed to obtain strategy optimization data, and periodic indicators are predicted through the strategy optimization data and plant morphological evolution data to form management guidance data; The quality index is characterized by analyzing the management guidance data and the plant maturity image data to obtain the harvest timing data, and the grade parameters are calculated according to the harvest timing data and the crop quality characteristic data to generate the quality assessment data.

2. The smart agricultural planting method according to claim 1, characterized in that: The method of performing a multi-dimensional analysis on growth indicators based on the nutrient ration data and seedling image feature data to obtain growth prediction data, calculating cultivation condition parameters based on the growth prediction data and leaf morphology feature data to generate seedling standardized data includes: The nutrient ration data is fused with the growth stage characteristics, the seedling stem thickness, internode distance, and leaf data are collected to form seedling image feature data, and the seedling image feature data is numerically counted to obtain growth benchmark parameters; Performing correlation analysis on the growth benchmark parameters and historical seedling data, extracting key growth characteristics, correlating the key growth characteristics with environmental parameters, constructing a growth indicator matrix, and obtaining growth prediction data through multidimensional data processing; Collecting seedling leaf images, extracting data on leaf color, area, and shape, generating leaf morphological feature data, matching the leaf morphological feature data with the growth prediction data, and obtaining growth status parameters; Dividing the growth state parameter into intervals, determining a cultivation condition threshold, correlating the cultivation condition threshold with environmental historical data to form a cultivation condition parameter, and performing data verification on the cultivation condition parameter; Performing time series processing on the cultivation condition parameters and the growth cycle data, calculating the seedling development index, and performing data correction on the seedling development index to form a standardized index; Data screening is performed on the standardized indicators, a seedling quality evaluation system is established, data comparison is performed between the seedling quality evaluation system and target parameters, and seedling standardized data is generated.

3. The smart agricultural planting method according to claim 1, characterized in that: The method of performing numerical analysis on environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and performing feature extraction and calculation on spectral parameters through the environmental control parameter data and plant canopy image data to form environmental regulation data includes: The nutrient ration data is coupled with the growth prediction data, environmental influencing factors are extracted, numerical calculations are performed on the environmental influencing factors to form an environmental condition matrix, and data is split on the environmental condition matrix to obtain environmental control parameter data; Perform correlation analysis on the moisture, temperature and humidity data, establish a parameter response table, perform numerical verification on the environmental control parameter data, form environmental benchmark data, perform time series processing on the environmental benchmark data, and obtain environmental change trends; Collecting multi-angle images of plant canopies, extracting canopy density data, correlating the canopy density data with light intensity data to generate plant canopy image data, performing feature calculation on the plant canopy image data, and obtaining photosynthetic parameters; The photosynthetic parameters are fused with the environmental spectrum data, the spectrum characteristic values ​​are extracted, the spectrum characteristic values ​​are numerically calculated to form spectrum parameters, the spectrum parameters are compared and analyzed with the historical data, and the spectrum regulation index is established; Performing data stratification on the spectral control index, calculating the spectral compensation coefficient, performing data matching between the spectral compensation coefficient and the environmental factor to form control strategy data, and performing numerical verification on the control strategy data; The control strategy data and the environmental control parameter data are processed comprehensively to construct an environmental response matrix, and data optimization is performed on the environmental response matrix to generate environmental control data.

4. The smart agricultural planting method according to claim 1, characterized in that: The method of dynamically analyzing the growth state based on the environmental control parameter data and the plant growth image sequence data to obtain growth evaluation data, and calculating the health index based on the growth evaluation data and the pest and disease image feature data to generate health diagnosis data includes: Perform time-series matching of the environmental control parameter data and the plant growth status data, extract key growth time points, collect images of plants at different growth stages, generate plant growth image sequence data, perform feature extraction on the plant growth image sequence data, and obtain growth change parameters; Compare and calculate the growth change parameters with standard growth data to establish a growth deviation matrix, perform numerical analysis on the growth deviation matrix to form a growth state index, perform data association on the growth state index to obtain growth assessment data; Collecting disease spot images on plant leaves and stems, extracting disease spot feature points, matching the disease spot feature points with standard disease data, generating disease and insect pest image feature data, classifying the disease and insect pest image feature data, and obtaining disease type parameters; fusing the growth assessment data with the disease type parameters to construct a health assessment matrix, performing numerical operations on the health assessment matrix to form health indicators, and comparing and analyzing the health indicators with historical data; Performing data stratification on the health indicators, calculating health risk coefficients, performing data association between the health risk coefficients and environmental factors to form diagnostic strategy data, and performing numerical verification on the diagnostic strategy data; The diagnostic strategy data and the growth assessment data are processed comprehensively to construct a health response matrix, and the health response matrix is ​​optimized to generate health diagnostic data.

5. The smart agricultural planting method according to claim 1, characterized in that: The time series analysis of growth parameters based on the growth assessment data and the health diagnosis data is performed to obtain strategy optimization data, and the periodic indicators are predicted through the strategy optimization data and the plant morphological evolution data to form management guidance data, including: Performing time series correlation on the growth assessment data and the health diagnosis data, extracting growth stage feature points, performing numerical statistics on the growth stage feature points to form a growth parameter matrix, performing time series analysis on the growth parameter matrix, and obtaining a growth fluctuation index; Compare and calculate the growth fluctuation index with the historical growth data, establish a parameter optimization matrix, perform numerical analysis on the parameter optimization matrix, form decision parameters, perform data verification on the decision parameters, and obtain strategy optimization data; Collecting images of plant morphological changes during growth, extracting morphological feature points, matching the morphological feature points with growth cycle data, generating plant morphological evolution data, classifying the plant morphological evolution data, and obtaining morphological type parameters; The strategy optimization data is integrated with the pattern type parameters to construct a cycle evaluation matrix, a numerical operation is performed on the cycle evaluation matrix to form a cycle index, and the cycle index is compared and analyzed with historical data; Performing data stratification on the periodic indicators, calculating the periodic adjustment coefficient, performing data association between the periodic adjustment coefficient and the environmental factor to form management strategy data, and performing numerical verification on the management strategy data; The management strategy data and the strategy optimization data are processed comprehensively to construct a management response matrix, and data optimization is performed on the management response matrix to generate management guidance data.

6. The smart agricultural planting method according to claim 1, characterized in that: The characteristic analysis of the quality index is performed based on the management guidance data and the plant maturity image data to obtain the harvest timing data, and the grade parameters are calculated based on the harvest timing data and the crop quality characteristic data to generate the quality assessment data, including: Performing time series matching of the management guidance data and the crop growth stage data, extracting maturity feature points, collecting images of plant fruits and seeds, generating plant maturity image data, performing feature extraction on the plant maturity image data, and obtaining maturity parameters; Compare and calculate the maturity parameters with the quality standard data to establish a quality assessment matrix, perform numerical analysis on the quality assessment matrix to form quality indicators, perform data correlation on the quality indicators, and obtain harvest timing data; Collecting images of crop appearance, size, and color, extracting quality feature points, matching the quality feature points with standard quality data, generating crop quality feature data, classifying the crop quality feature data, and obtaining quality type parameters; fusing the harvesting time data with the quality type parameters to construct a grade evaluation matrix, performing numerical operations on the grade evaluation matrix to form grade parameters, and comparing and analyzing the grade parameters with historical data; Performing data stratification on the grade parameters, calculating the comprehensive quality coefficient, performing data association between the comprehensive quality coefficient and the market factor to form evaluation strategy data, and performing numerical verification on the evaluation strategy data; The evaluation strategy data and the harvesting timing data are processed comprehensively to construct a quality response matrix, and data optimization is performed on the quality response matrix to generate quality evaluation data.

7. The smart agricultural planting method according to claim 6, characterized in that: The step of stratifying the grade parameters, calculating the comprehensive quality coefficient, associating the comprehensive quality coefficient with market factors to form evaluation strategy data, and numerically verifying the evaluation strategy data includes: Numerically sorting the grade parameters, establishing a quality hierarchy matrix, dividing the quality hierarchy matrix into intervals, generating quality stratified data, performing threshold processing on the quality stratified data, and obtaining stratified boundary points; Performing correlation analysis on the stratified boundary points and historical quality data, establishing a quality evaluation matrix, performing feature calculation on the quality evaluation matrix, forming a quality weight parameter, and numerically verifying the quality weight parameter to obtain a comprehensive quality coefficient; Collect market price, supply and demand, and seasonal data, extract key market features, match the key market features with historical transaction data, generate market factor data, perform trend analysis on the market factor data, and obtain market volatility parameters; Data coupling is performed on the comprehensive quality coefficient and the market volatility parameter to construct a strategy calculation matrix, numerical analysis is performed on the strategy calculation matrix to form strategy indicators, and data screening is performed on the strategy indicators; Performing time series processing on the strategy indicators, calculating strategy adjustment coefficients, fusing the strategy adjustment coefficients with environmental factors to form evaluation strategy data, and comparing the evaluation strategy data with benchmark data; The reliability of the evaluation strategy data is calculated, a verification index matrix is ​​constructed, data optimization is performed on the verification index matrix, and the evaluation strategy data is confirmed.

8. The intelligent agricultural planting method according to claim 7, characterized in that: The time series processing of the strategy indicator, calculating the strategy adjustment coefficient, fusing the strategy adjustment coefficient with environmental factors to form evaluation strategy data, and comparing the evaluation strategy data with benchmark data include: Performing time series analysis on the strategy indicators, extracting trend feature points, performing statistical processing on the trend feature points, generating time series variation data, performing period analysis on the time series variation data, and obtaining fluctuation law parameters; Calculate the correlation between the fluctuation law parameter and the historical strategy data, establish a strategy response matrix, extract features from the strategy response matrix to form an adjustment weight parameter, perform numerical calculation on the adjustment weight parameter, and obtain a strategy adjustment coefficient; Collect temperature, humidity, and light intensity data, extract environmental characteristic values, match the environmental characteristic values ​​with historical environmental data, generate environmental factor data, perform correlation analysis on the environmental factor data, and obtain environmental impact parameters; Integrate the strategy adjustment coefficient and the environmental impact parameter to construct a fusion calculation matrix, perform numerical operations on the fusion calculation matrix to form a fusion index, and perform data verification on the fusion index; Optimizing the fusion index, calculating the evaluation coefficient, performing feature matching between the evaluation coefficient and historical data to form evaluation strategy data, and numerically verifying the evaluation strategy data; The evaluation strategy data is compared with the planting benchmark data to construct a comparison matrix, and a numerical analysis is performed on the comparison matrix to generate the evaluation strategy data.

9. An intelligent agricultural planting system, used to implement the intelligent agricultural planting method according to any one of claims 1 to 8, characterized in that: The intelligent agricultural planting system comprises: The acquisition module is used to analyze and process the nutritional requirements according to the environmental sensor data and the plant image data to obtain the nutritional ratio parameter data, extract and calculate the characteristics of the nutrient formula through the nutritional ratio parameter data, and generate the nutritional ration data; An analysis module, for performing multi-dimensional analysis on growth indicators based on the nutrient ration data and seedling image feature data to obtain growth prediction data, calculating cultivation condition parameters based on the growth prediction data and leaf morphology feature data to generate seedling standardized data; An extraction module is used to perform numerical analysis on environmental conditions according to the nutrient ration data and the growth prediction data to obtain environmental control parameter data, and to perform feature extraction and calculation on spectral parameters through the environmental control parameter data and plant canopy image data to form environmental regulation data; A calculation module, used to dynamically analyze the growth status according to the environmental control parameter data and the plant growth image sequence data to obtain growth evaluation data, calculate the health index according to the growth evaluation data and the pest and disease image feature data, and generate health diagnosis data; A prediction module, used to perform time series analysis on growth parameters based on the growth assessment data and the health diagnosis data to obtain strategy optimization data, and to predict periodic indicators through the strategy optimization data and plant morphological evolution data to form management guidance data; The evaluation module is used to perform feature analysis on quality indicators based on the management guidance data and the plant maturity image data to obtain harvest timing data, calculate grade parameters based on the harvest timing data and the crop quality feature data, and generate quality evaluation data.

Images