Plateau region-based summer vegetable planting method and system

Abstract

The invention relates to the technical field of intelligent agriculture, and discloses a planting method and system based on summer vegetables in a plateau area. The method comprises the following steps: continuously acquiring soil and meteorological multi-parameter data through a sensor array deployed in a target planting area, and performing verification processing to form a standardized data set; carrying out planting suitability analysis by using the data set to generate a spatial distribution map; establishing a crop growth prediction model in combination with historical data and real-time data, and outputting growth expectations of different planting schemes; formulating a detailed planting plan including time, density and position according to the map and an expected result; controlling the planting equipment to sow according to the plan, and continuously monitoring the growth vigor; and dynamically adjusting management measures according to the growth vigor change to form closed-loop regulation and control. According to the plateau summer vegetable planting management system, refining of planting planning and intellectualization of growth management are achieved, and the resource utilization efficiency and output stability of plateau summer vegetable planting are improved.

Description

Technical Field

[0001] This invention relates to the field of smart agriculture technology, specifically to a method and system for planting summer vegetables in high-altitude regions. Background Technology

[0002] Summer vegetable cultivation in highland areas relies heavily on traditional farming experience and fixed planting patterns. During the planning stage, farmers typically make decisions based on their overall impression of the fields and historical habits, lacking a precise understanding of the spatial variability of key environmental factors such as soil moisture, nutrients, and temperature within the planting area. Existing technologies usually employ a combination of limited-point sampling and manual surveys to assess land conditions. This method yields sparse and outdated data, making it difficult to form a refined and quantitative spatial understanding of the entire planting area. This results in significant uncertainty in the selection of planting locations and variety combinations, failing to achieve optimal spatial allocation of agricultural resources.

[0003] In crop growth management, existing technologies primarily rely on pre-set, standardized agronomic calendars for irrigation, fertilization, and other operations, resulting in static and generic management strategies. Due to the significant microclimate characteristics and complex, variable weather conditions in high-altitude regions, this fixed management model struggles to adapt to the actual needs of crops at different growth stages and in different plots. The application of crop growth models largely remains at the research or post-event analysis level, failing to deeply integrate with real-time environmental monitoring and agricultural operations, and thus unable to provide forward-looking simulations of different planting schemes to support decision-making. Management adjustments during the growth process also heavily depend on manual inspections and experience-based judgment, leading to untimely responses and a lack of quantitative data. The entire planting process has not yet formed an optimization loop based on data feedback. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for cultivating summer vegetables in high-altitude areas, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for cultivating summer vegetables in high-altitude regions, the method comprising:

[0006] By continuously collecting multi-parameter soil data and multi-parameter meteorological data through sensor arrays deployed in the target planting area, the collected multi-parameter soil data and meteorological data are subjected to quality verification and outlier removal to form a standardized environmental monitoring dataset.

[0007] The standardized environmental monitoring dataset is used to perform planting suitability analysis, generating a planting suitability map that includes spatial distribution characteristics;

[0008] By combining historical planting records with real-time environmental monitoring data, a crop growth prediction model is established, and the expected growth results under different planting schemes are output through the crop growth prediction model.

[0009] Based on the planting suitability map and expected growth results, a detailed planting operation plan is formulated, which includes planting time arrangement, planting density planning and planting location allocation;

[0010] The planting equipment is controlled to perform sowing operations according to the planting operation plan, and crop growth changes are continuously monitored during the crop growth cycle;

[0011] Management measures are dynamically adjusted based on monitored changes in crop growth to form a closed-loop planting control strategy.

[0012] Preferably, the continuous collection of soil multi-parameter data and meteorological multi-parameter data through a sensor array deployed in the target planting area specifically includes:

[0013] Soil sensor nodes and meteorological sensor nodes are arranged in the target planting area according to a predetermined grid pattern. The soil sensor nodes measure soil moisture, soil temperature and soil pH, while the meteorological sensor nodes measure air temperature, air humidity and light intensity.

[0014] Set the data acquisition frequency, and each sensor node synchronously collects environmental parameters according to the set frequency, and sends the collected data to the data aggregation node through the wireless sensor network;

[0015] The data aggregation node performs timestamp alignment and format unification processing on the received sensor data to form the original environmental dataset;

[0016] Data cleaning is performed on the original environmental dataset to identify and remove abnormal data points caused by sensor malfunctions or transmission interference.

[0017] Spatial interpolation is performed on the cleaned environmental data to generate a continuous environmental parameter distribution map covering the entire target planting area.

[0018] Preferably, the step of using the standardized environmental monitoring dataset to perform planting suitability analysis and generate a planting suitability map containing spatial distribution characteristics specifically includes:

[0019] Spatial distribution data of various environmental parameters are extracted from standardized environmental monitoring datasets to establish a spatial database of environmental parameters.

[0020] Determine the suitable range thresholds for various environmental parameters for different summer vegetable varieties, and construct a crop environmental suitability evaluation standard;

[0021] The spatial overlay analysis method was used to match and analyze the spatial distribution data of various environmental parameters with the crop environmental suitability evaluation criteria.

[0022] Calculate the degree of conformity between the environmental parameters of each spatial location and the suitable range, and generate a suitability score layer for each environmental parameter;

[0023] Weighted fusion of the suitability score layers for each environmental parameter was performed to obtain a comprehensive planting suitability distribution map.

[0024] Spatially smooth the comprehensive planting suitability distribution map to eliminate local abnormal areas and form the final planting suitability map.

[0025] Preferably, the step of establishing a crop growth prediction model by combining historical planting record data with real-time environmental monitoring data specifically includes:

[0026] Collect historical planting records for the target planting area over multiple growth cycles, including planting time, planting varieties, environmental data, and final yield data;

[0027] Organize and standardize historical planting records to establish a historical planting database;

[0028] Analyze the correlation between environmental parameters and crop growth indicators in historical planting databases to identify key influencing factors;

[0029] We chose neural networks as the basic architecture of the model and constructed a growth prediction network structure that includes an input layer, hidden layers, and an output layer.

[0030] The growth prediction network was trained using a historical planting database, and the network weight parameters were adjusted to minimize the error between the prediction output and the actual growth data.

[0031] Real-time environmental monitoring data is input into the trained growth prediction model to obtain the expected growth results under different planting schemes.

[0032] Preferably, the step of formulating a detailed planting operation plan based on the planting suitability map and expected growth results specifically includes:

[0033] The distribution of high-suitability and low-suitability areas was determined based on the planting suitability map;

[0034] Based on the expected yield data of different regions in the growth forecast results, the planting priority of each region is determined;

[0035] Considering crop rotation requirements and soil restoration needs, planting time sequences should be rationally arranged for different regions;

[0036] Based on the characteristics of the planted varieties and their expected growth conditions, determine the planting density and planting methods for each region;

[0037] By taking into account spatial distribution, time arrangement, and planting parameters, a planting operation plan containing specific operational instructions is formed;

[0038] The planting operation plan is broken down into an actionable task list, with clear timeframes and execution requirements for each task.

[0039] Preferably, the controlled planting equipment performs sowing operations according to the planting operation plan and continuously monitors crop growth changes during the crop growth cycle, specifically including:

[0040] Convert the planting operation plan into a sequence of control commands that the planting equipment can recognize;

[0041] The planting equipment automatically controls the sowing depth, sowing spacing, and sowing quantity according to the sequence of control commands.

[0042] After sowing, start the multispectral imaging equipment to collect crop canopy image data regularly;

[0043] Image processing techniques are used to extract crop growth parameters such as leaf area index, chlorophyll content, and plant height from canopy images.

[0044] Establish a crop growth time series database to record the growth trajectory of crops throughout their entire growth cycle;

[0045] When abnormal changes in growth parameters are detected, an early warning mechanism is automatically triggered and the abnormal situation is recorded.

[0046] Preferably, the planting control strategy of dynamically adjusting management measures based on monitored crop growth changes to form a closed loop specifically includes:

[0047] Set the normal range threshold for growth parameters of crops at each growth stage;

[0048] By comparing and analyzing the real-time monitored crop growth parameters with the normal range threshold, deviations in growth can be identified.

[0049] Based on the degree and type of growth deviation, a targeted control plan is automatically generated, including irrigation adjustments, fertilization adjustments, and pest and disease control measures.

[0050] The control plan is transformed into specific equipment control instructions, which automatically adjust the working parameters of the irrigation system and fertilization device.

[0051] After implementing control measures, continue to monitor crop growth changes and assess the effectiveness of the control measures;

[0052] Based on feedback on the control effect, the control strategy parameters are optimized to form an adaptive closed-loop control system.

[0053] Preferably, the step of extracting crop leaf area index, chlorophyll content, and plant height from canopy images using image processing technology specifically includes:

[0054] The acquired multispectral canopy images were preprocessed, including radiometric calibration, atmospheric correction, and geometric correction.

[0055] An image segmentation algorithm was used to separate the crop canopy from the background soil, and the pure crop canopy region was extracted.

[0056] Calculate the spectral reflectance of the canopy region in different bands and construct the spectral characteristic curve of the crop;

[0057] Leaf area index was retrieved by analyzing spectral characteristics, and normalized vegetation index was calculated using the reflectance ratio of near-infrared band to red band.

[0058] Chlorophyll content index was calculated based on red-edge band features, and plant height was measured using stereoscopic vision technology;

[0059] The extracted growth parameters are correlated with the corresponding location information to generate a spatialized growth distribution map.

[0060] Preferably, the weighted fusion of the suitability score layers for each environmental parameter to obtain a comprehensive planting suitability distribution map specifically includes:

[0061] Based on the growth requirements of the target summer vegetable varieties, the weight coefficients of each environmental parameter are determined by the analytic hierarchy process. The environmental parameters include soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity.

[0062] The suitability score layers for each environmental parameter are converted into raster data with the same spatial resolution to ensure that each raster cell corresponds to a consistent geographical location.

[0063] For each grid cell, the sum of the products of the suitability score of each environmental parameter and the corresponding weight coefficient is calculated to obtain the weighted suitability score of the grid cell.

[0064] The weighted suitability scores of all grid cells are linearly normalized to map the scores to a range of 0 to 1, generating a comprehensive planting suitability distribution map.

[0065] To verify the accuracy of the integrated planting suitability distribution map, consistency was checked by comparing the suitability scores of historically successful planting areas.

[0066] Preferably, the present invention also includes a planting system for summer vegetables in high-altitude areas, the system including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of the planting method for summer vegetables in high-altitude areas as described above.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] By deploying sensor arrays and generating planting suitability maps containing spatial distribution characteristics, discrete point-based environmental monitoring data is transformed into continuous spatial visualization information. This allows for the precise quantification of microenvironmental differences within planting areas, revealing the specific distribution patterns of soil and climate conditions in the field. Based on this map, planting operation plans can be developed to accurately allocate the most suitable crop varieties to the most appropriate plots, achieving spatial optimization of planting location and variety layout. This improves land resource utilization efficiency, avoids resource waste and yield reduction risks caused by blindly planting in unsuitable areas, and creates optimal starting conditions for crop growth from a spatial planning perspective.

[0069] By combining historical and real-time data to establish a growth prediction model and form a closed-loop control strategy, pre-planting plan assessment and in-process optimization of the growth process were achieved. Before planting, the model simulated the expected results of different planting plans, providing quantitative decision support for selecting key parameters such as optimal sowing time and density, reducing the randomness of production. During the growth cycle, continuous monitoring of crop growth and comparison with prediction results enabled timely identification of growth deviations. Based on this feedback, water and fertilizer management measures were dynamically adjusted, transforming management behavior from a fixed program to dynamic and precise intervention in response to the actual needs of the crop. This enhanced the predictability and adaptability of the planting management process, effectively coped with the variable characteristics of the plateau environment, promoted crop growth towards the expected goals, and stabilized and improved the final output benefits. Attached Figure Description

[0070] Figure 1 This is a schematic diagram illustrating the working principle of the summer vegetable cultivation method based on plateau regions described in this invention.

[0071] Figure 2 A flowchart for soil and meteorological data acquisition and processing;

[0072] Figure 3 A flowchart for generating a planting suitability map;

[0073] Figure 4 Bar chart showing the weight distribution of environmental parameters suitable for highland summer vegetable cultivation.

[0074] Figure 5 Bar graph showing the correlation between environmental parameters and crop yield of summer vegetables in highland areas. Detailed Implementation

[0075] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] Please see Figure 1 This invention provides a method for cultivating summer vegetables in high-altitude areas. The method includes: continuously collecting multi-parameter soil data and multi-parameter meteorological data using a sensor array deployed in the target planting area; performing quality verification and outlier removal on the collected soil and meteorological data to form a standardized environmental monitoring dataset; using this standardized environmental monitoring dataset to perform planting suitability analysis and generate a planting suitability map containing spatial distribution characteristics; establishing a crop growth prediction model by combining historical planting records and real-time environmental monitoring data, and outputting the expected growth results under different planting schemes through this crop growth prediction model; formulating a detailed planting operation plan based on the planting suitability map and the expected growth results, which includes planting time arrangement, planting density planning, and planting location allocation; controlling planting equipment to perform sowing operations according to the planting operation plan, and continuously monitoring crop growth changes throughout the crop growth cycle; and dynamically adjusting management measures based on the monitored crop growth changes to form a closed-loop planting control strategy.

[0077] Example 1: See Figure 2 In practice, the target planting area is divided into a regular grid pattern for sensor deployment. The grid size is determined based on the planting area and monitoring accuracy requirements, for example, a 20-meter by 20-meter grid spacing. Soil sensor nodes and meteorological sensor nodes are installed at each grid point. The soil sensor nodes are buried 20 centimeters below the surface to continuously measure soil moisture, soil temperature, and soil pH parameters. The meteorological sensor nodes are fixed on a support 1.5 meters above the ground to continuously measure air temperature, air humidity, and light intensity parameters. The data acquisition frequency is set to once every 30 minutes. Each sensor node has a built-in synchronization clock module to ensure consistent acquisition time. The collected environmental parameters are transmitted to the data aggregation node via a wireless sensor network using a low-power wide-area network protocol. After receiving data packets from each sensor, the data aggregation node performs timestamp alignment to unify the data from different nodes into the same time coordinate system and performs data format standardization processing to form a raw environmental dataset containing time series and spatial location.

[0078] In some embodiments, a data cleaning process is performed on the original environmental dataset, employing a statistical anomaly detection method, such as calculating the Z-score for each parameter value. Data points with an absolute Z-score greater than 3 are identified as outliers and removed. The cleaning process is automatically iterated until no new outliers are added. It can be understood that spatial interpolation is performed on the cleaned environmental data, using an inverse distance weighted interpolation algorithm to generate a continuous distribution map. The interpolation formula is expressed as:

[0079]

[0080] in: These are the environmental parameter values ​​for the interpolation point (x, y). It is the number of known data points in the surrounding area. It is the parameter value of the i-th known point. It is the Euclidean distance between point (x, y) and the i-th known point. The distance attenuation power parameter is typically set to 2. The interpolation result generates a raster-style environmental parameter distribution map with a raster resolution consistent with the sensor grid spacing, fully covering the target planting area.

[0081] Optionally, the wireless sensor network is configured with self-organizing network capabilities, automatically routing data to the aggregation node when individual nodes fail. In some embodiments, the data aggregation node integrates a data caching mechanism to handle network transmission interruptions. It is understood that a normalization operation is performed on the data before spatial interpolation to eliminate dimensional differences.

[0082] Example 2: See Figure 3 In practice, spatial distribution data of various environmental parameters, including soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity, are extracted from standardized environmental monitoring datasets. This data is stored in raster form to establish a spatial database of environmental parameters, with each raster cell associated with specific geographic coordinates. Suitable range thresholds for each environmental parameter are determined for different summer vegetable varieties. For example, the suitable range threshold for soil moisture for a certain variety is set at 60% to 80% of field capacity, and the suitable range threshold for soil pH is set at pH 6.0 to 7.0. Based on these thresholds, crop environmental suitability evaluation standards are constructed. Spatial overlay analysis is used to match the spatial distribution data of each environmental parameter with the crop environmental suitability evaluation standards, and the suitability of the environmental parameter values ​​for each raster cell is assessed.

[0083] In some embodiments, the degree of conformity between environmental parameters at each spatial location and the suitable range is calculated. A piecewise function is used to calculate the suitability score of each environmental parameter. For each grid cell, soil moisture suitability score, soil temperature suitability score, soil pH suitability score, air temperature suitability score, air humidity suitability score, and light intensity suitability score layers are generated. Based on the growth requirements of the target summer vegetable variety, the weight coefficients of each environmental parameter are determined using the analytic hierarchy process (AHP). A judgment matrix is ​​constructed for consistency verification, and finally, the weight coefficients of soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity are obtained.

[0084] Optionally, a weighted fusion calculation is performed on each grid cell, expressed by the formula:

[0085]

[0086] in: This represents the overall planting suitability score of the j-th grid cell. This represents the weight coefficient of the i-th environmental parameter. This represents the suitability score of the i-th environmental parameter for the j-th grid cell. The values ​​range from 1 to 6, corresponding to six environmental parameters: soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity. The comprehensive planting suitability score for all grid cells is linearly normalized, mapping the score values ​​to a range of 0 to 1, generating a comprehensive planting suitability distribution map. In some embodiments, the accuracy of the comprehensive planting suitability distribution map is verified by comparing the suitability scores of historically successful planting areas for consistency checks. The average suitability score of historically high-yield planting areas is calculated as the verification benchmark. This can be understood as spatially smoothing the comprehensive planting suitability distribution map, using a moving window averaging method to eliminate local anomalies, with the window size set to a 3×3 grid, forming the final planting suitability map. Optionally, the suitability map is visualized as a color-rendered image, with different colors used to label different suitability levels.

[0087] See Figure 4 This chart is a visualization of the weighted fusion of environmental parameter suitability, presenting the suitability scores and weights of six core environmental parameters for highland summer vegetable cultivation in a bar chart format. The parameters covered in the chart include soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity, with weights determined using the analytic hierarchy process (AHP). This quantitative display makes the impact of environmental factors on planting suitability more intuitive, serving as a crucial intermediate step from discrete environmental data to refined planting decisions, and embodying the core logic of data quantification supporting spatial optimization in smart agriculture.

[0088] Example 3: In the specific implementation, historical planting records from multiple growth cycles in the target planting area were collected. These records included planting time information, crop variety information, environmental data, and final yield information. Environmental data covered soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity. The historical planting records were organized and standardized to eliminate differences in data recording formats across different years, unify data units and time scales, and establish a structured historical planting database. The correlation between environmental parameters and crop growth indicators in the historical planting database was analyzed. The Pearson correlation coefficient method was used to quantify the impact of each environmental parameter on crop yield and identify key influencing factors. In analyzing the correlation between environmental parameters and crop growth indicators in the historical planting database, time-series data of each environmental parameter and corresponding crop yield data were extracted from the database. Environmental parameters included soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity. Crop growth indicators were expressed as yield per unit area. The Pearson correlation coefficient between each environmental parameter and crop yield is calculated. The process includes calculating the covariance and standard deviation of the environmental parameter data series and the yield data series, and then obtaining the correlation coefficient value by the ratio. The correlation coefficient ranges from -1 to 1, with positive values ​​indicating positive correlation and negative values ​​indicating negative correlation.

[0089] In some embodiments, a neural network is selected as the model architecture to construct a growth prediction network structure containing an input layer, hidden layers, and an output layer. The number of nodes in the input layer corresponds to the number of key influencing factors, the hidden layer adopts a two-layer fully connected structure, and the output layer corresponds to crop growth indicators. The growth prediction network is trained using a historical planting database. The historical planting records are divided into training and validation sets in chronological order. The network weight parameters are adjusted using the backpropagation algorithm, and the mean squared error is used as the loss function during training.

[0090] Optionally, the neural network training employs an adaptive moment estimation algorithm for optimization, with a learning rate set to 0.001 and a batch size of 32. Real-time environmental monitoring data is input into the trained growth prediction model. The model processing flow includes three stages: data preprocessing, forward propagation computation, and result output. In some embodiments, the data preprocessing stage standardizes the real-time environmental monitoring data to the model input format requirements, and the forward propagation computation stage performs weighted summations and activation function transformations of each layer of the network. The mathematical expression of the growth prediction model is as follows:

[0091]

[0092] in: This represents the input environment parameter vector. This indicates the output of the predicted growth index. and These represent the weight matrices of the hidden layer and the output layer, respectively. and This represents the bias vector. and This represents the activation function. It can be understood that growth prediction models can obtain expected growth results under different planting schemes, including key indicators such as yield prediction and growth cycle prediction.

[0093] See Figure 5 This figure visualizes the correlation analysis between environmental parameters and crop growth indicators, primarily showcasing the correlation coefficients between six planting environmental parameters and summer vegetable yield. The value of this figure lies in its quantification of the correlation between environmental factors and yield from historical planting databases. It serves as a direct basis for selecting key influencing factors for growth prediction models. When subsequently building neural network models, these highly correlated parameters will be prioritized as input layer nodes, avoiding model redundancy and improving prediction accuracy. This step is a crucial link between historical data processing and growth prediction model construction, reflecting the core logic of data correlation analysis supporting quantitative decision-making in smart agriculture.

[0094] Example 4: In specific implementation, the planting operation plan is formulated based on the planting suitability map and the expected growth results. The planting suitability map provides a spatial distribution of suitability scores, and high-suitability areas and low-suitability areas are distinguished by setting thresholds. For example, a high-suitability area is defined as an area with a suitability score greater than 0.7, and a low-suitability area is defined as an area with a suitability score less than 0.4. Combined with the expected yield data of different areas in the expected growth results (in kilograms per hectare), a weighted average method is used to calculate the planting priority of each area. The priority calculation formula is expressed as:

[0095]

[0096] in: This represents the planting priority score for the m-th region. This represents the suitability score for the m-th region. This represents the normalized expected output value for the m-th region. and Indicates the weighting coefficient. The value is 0.5. A value of 0.5 is used to balance suitability and yield factors. After calculation, regions are sorted in descending order of priority score. In some embodiments, referring to Table 1, the priority assessment results are recorded in Table 1 for visualization reference.

[0097] Table 1: Priority Allocation Table for Planting Areas

[0098] Area Identification Suitability score Expected output (kg / ha) <![CDATA[Normalized yield P' m > Priority rating R001 0.82 4200 0.88 0.85 R002 0.65 3800 0.79 0.72 R003 0.91 4600 0.96 0.94

[0099] It is understandable that, considering crop rotation requirements and soil restoration needs, planting timelines should follow the crop rotation cycle principle. For example, areas where leafy vegetables were planted last year should be planted with root and tuber crops this year, with an interval of at least 30 days. Based on the characteristics of the planted varieties and expected growth conditions, planting density planning is based on recommended plant and row spacing values. In areas where expected growth results indicate high yields, the density should be appropriately increased. For example, if the standard density for a certain variety is 25,000 plants per hectare, it can be adjusted to 28,000 plants per hectare in high-priority areas.

[0100] Optionally, the planting method is determined based on soil texture and variety characteristics. For example, direct seeding is used in loose soil, while transplanting is used in heavy clay soil. Taking into account spatial distribution, time schedule, and planting parameters, the planting operation plan generates specific operational instructions, including the planting start date, planting machinery configuration parameters, and seed quantity for each area. The planting operation plan is broken down into an executable task list, specifying the task number, corresponding area, operation type, planned start time, planned completion time, and resource allocation details.

[0101] In some embodiments, the task list tracks progress using project management software. It is understood that timeline settings are compatible with weather forecast data to avoid the impact of severe weather. Optionally, the planting operation plan is output in a standardized file format for easy parsing and execution by the planting equipment system.

[0102] Example 5: In specific implementation, the planting operation plan is converted into a sequence of control instructions recognizable by the planting equipment. This sequence is encapsulated in JSON format and includes parameters for sowing depth control, sowing spacing adjustment, and sowing quantity control. The planting equipment's controller parses the instructions and automatically executes the corresponding operations. Sowing depth control is achieved by adjusting the depth of the furrow opener in the soil using a hydraulic system; sowing spacing adjustment is achieved by adjusting the speed of the seed metering device; and sowing quantity control is achieved by precisely controlling the seed flow rate using a metering device. After the sowing operation is completed, a high-resolution multispectral imaging device deployed in the field is activated. The multispectral imaging device collects crop canopy image data at a preset cycle, set to 7 days.

[0103] The acquired multispectral canopy images underwent preprocessing, including radiometric calibration, atmospheric correction, and geometric correction. Radiometric calibration converted the image digital values ​​into surface reflectance, atmospheric correction eliminated aerosol effects, and geometric correction eliminated image distortion. An image segmentation algorithm was used to separate the crop canopy from the background soil. This algorithm, based on color space transformation and thresholding, extracted pure crop canopy regions for subsequent analysis. The spectral reflectance of the canopy region in different wavelength bands was calculated, constructing crop spectral characteristic curves. These curves included reflectance values ​​in the blue, green, red, red-edge, and near-infrared bands.

[0104] Leaf area index was retrieved through spectral feature analysis, and the normalized vegetation index (NDI) was calculated using the ratio of reflectance in the near-infrared band to the red band. The formula for calculating the NDI is as follows:

[0105]

[0106] in: This represents the normalized vegetation index value. Indicates near-infrared reflectivity. The red band reflectance is represented. The chlorophyll content index is calculated based on the red-edge band characteristics. Plant height is measured using stereo vision technology, with the stereo vision system acquiring three-dimensional point cloud data of the canopy using dual cameras. The extracted leaf area index, chlorophyll content, and plant height, among other growth parameters, are correlated with corresponding geographic coordinates to generate a spatialized growth distribution map.

[0107] A crop growth time-series database is established to record the growth trajectory of crops throughout their entire growth cycle. The database stores the growth parameter values ​​of each monitoring point by date. When abnormal changes in growth parameters are detected, an early warning mechanism is automatically triggered and the anomaly is recorded. The early warning mechanism is set based on the standard deviation multiple of the parameter from the normal range. Normal range thresholds for growth parameters at each growth stage are set; for example, the leaf area index threshold range is 0.5-1.2 during the seedling stage and 1.5-3.0 during the mid-growth stage.

[0108] The system compares and analyzes real-time monitored crop growth parameters with normal threshold ranges to identify growth deviations, categorizing them into three levels: slight, moderate, and severe. Based on the degree and type of deviation, it automatically generates targeted control plans, including irrigation adjustments, fertilization adjustments, and pest and disease control measures. These plans are then translated into specific equipment control commands to automatically adjust the operating parameters of the irrigation system and fertilization devices, such as increasing irrigation duration or adjusting fertilizer ratios.

[0109] After implementing control measures, crop growth changes are continuously monitored to assess the effectiveness of the control. The assessment method uses the rate of change of growth parameters before and after control to calculate the effect. Based on the feedback of the control effect, the control strategy parameters are optimized to form an adaptive closed-loop control system. The adaptive closed-loop control system continuously adjusts the thresholds and control parameters through machine learning algorithms.

[0110] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for cultivating summer vegetables in high-altitude regions, characterized in that, Includes the following steps: By continuously collecting multi-parameter soil data and multi-parameter meteorological data through sensor arrays deployed in the target planting area, the collected multi-parameter soil data and meteorological data are subjected to quality verification and outlier removal to form a standardized environmental monitoring dataset. The standardized environmental monitoring dataset is used to perform planting suitability analysis, generating a planting suitability map that includes spatial distribution characteristics; By combining historical planting records with real-time environmental monitoring data, a crop growth prediction model is established, and the expected growth results under different planting schemes are output through the crop growth prediction model. Based on the planting suitability map and expected growth results, a detailed planting operation plan is formulated, which includes planting time arrangement, planting density planning and planting location allocation; The planting equipment is controlled to perform sowing operations according to the planting operation plan, and crop growth changes are continuously monitored during the crop growth cycle; Management measures are dynamically adjusted based on monitored changes in crop growth to form a closed-loop planting control strategy.

2. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The continuous collection of multi-parameter soil data and multi-parameter meteorological data through a sensor array deployed in the target planting area specifically includes: Soil sensor nodes and meteorological sensor nodes are arranged in the target planting area according to a predetermined grid pattern. The soil sensor nodes measure soil moisture, soil temperature and soil pH, while the meteorological sensor nodes measure air temperature, air humidity and light intensity. Set the data acquisition frequency, and each sensor node synchronously collects environmental parameters according to the set frequency, and sends the collected data to the data aggregation node through the wireless sensor network; The data aggregation node performs timestamp alignment and format unification processing on the received sensor data to form the original environmental dataset; Data cleaning is performed on the original environmental dataset to identify and remove abnormal data points caused by sensor malfunctions or transmission interference. Spatial interpolation is performed on the cleaned environmental data to generate a continuous environmental parameter distribution map covering the entire target planting area.

3. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The step of using the standardized environmental monitoring dataset to perform planting suitability analysis and generate a planting suitability map containing spatial distribution characteristics specifically includes: Spatial distribution data of various environmental parameters are extracted from standardized environmental monitoring datasets to establish a spatial database of environmental parameters. Determine the suitable range thresholds for various environmental parameters for different summer vegetable varieties, and construct a crop environmental suitability evaluation standard; The spatial overlay analysis method was used to match and analyze the spatial distribution data of various environmental parameters with the crop environmental suitability evaluation criteria. Calculate the degree of conformity between the environmental parameters of each spatial location and the suitable range, and generate a suitability score layer for each environmental parameter; Weighted fusion of the suitability score layers for each environmental parameter was performed to obtain a comprehensive planting suitability distribution map. Spatially smooth the comprehensive planting suitability distribution map to eliminate local abnormal areas and form the final planting suitability map.

4. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The establishment of a crop growth prediction model by combining historical planting records and real-time environmental monitoring data specifically includes: Collect historical planting records for the target planting area over multiple growth cycles, including planting time, planting varieties, environmental data, and final yield data; Organize and standardize historical planting records to establish a historical planting database; Analyze the correlation between environmental parameters and crop growth indicators in historical planting databases to identify key influencing factors; We chose neural networks as the basic architecture of the model and constructed a growth prediction network structure that includes an input layer, hidden layers, and an output layer. The growth prediction network was trained using a historical planting database, and the network weight parameters were adjusted to minimize the error between the prediction output and the actual growth data. Real-time environmental monitoring data is input into the trained growth prediction model to obtain the expected growth results under different planting schemes.

5. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The detailed planting operation plan, formulated based on the planting suitability map and expected growth results, specifically includes: The distribution of high-suitability and low-suitability areas was determined based on the planting suitability map; Based on the expected yield data of different regions in the growth forecast results, the planting priority of each region is determined; Considering crop rotation requirements and soil restoration needs, planting time sequences should be rationally arranged for different regions; Based on the characteristics of the planted varieties and their expected growth conditions, determine the planting density and planting methods for each region; By taking into account spatial distribution, time arrangement, and planting parameters, a planting operation plan containing specific operational instructions is formed; The planting operation plan is broken down into an actionable task list, with clear timeframes and execution requirements for each task.

6. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The controlled planting equipment performs sowing operations according to the planting operation plan and continuously monitors crop growth changes throughout the crop growth cycle, specifically including: Convert the planting operation plan into a sequence of control commands that the planting equipment can recognize; The planting equipment automatically controls the sowing depth, sowing spacing, and sowing quantity according to the sequence of control commands. After sowing, start the multispectral imaging equipment to collect crop canopy image data regularly; Image processing techniques are used to extract crop growth parameters such as leaf area index, chlorophyll content, and plant height from canopy images. Establish a crop growth time series database to record the growth trajectory of crops throughout their entire growth cycle; When abnormal changes in growth parameters are detected, an early warning mechanism is automatically triggered and the abnormal situation is recorded.

7. The method for cultivating summer vegetables in highland areas according to claim 1, characterized in that, The aforementioned planting control strategy, which dynamically adjusts management measures based on monitored crop growth changes to form a closed loop, specifically includes: Set the normal range threshold for growth parameters of crops at each growth stage; By comparing and analyzing the real-time monitored crop growth parameters with the normal range threshold, deviations in growth can be identified. Based on the degree and type of growth deviation, a targeted control plan is automatically generated, including irrigation adjustments, fertilization adjustments, and pest and disease control measures. The control plan is transformed into specific equipment control instructions, which automatically adjust the working parameters of the irrigation system and fertilization device. After implementing control measures, continue to monitor crop growth changes and assess the effectiveness of the control measures; Based on feedback on the control effect, the control strategy parameters are optimized to form an adaptive closed-loop control system.

8. A method for cultivating summer vegetables in high-altitude areas according to claim 6, characterized in that, The extraction of crop growth parameters such as leaf area index, chlorophyll content, and plant height from canopy images using image processing technology specifically includes: The acquired multispectral canopy images were preprocessed, including radiometric calibration, atmospheric correction, and geometric correction. An image segmentation algorithm was used to separate the crop canopy from the background soil, and the pure crop canopy region was extracted. Calculate the spectral reflectance of the canopy region in different bands and construct the spectral characteristic curve of the crop; Leaf area index was retrieved by analyzing spectral characteristics, and normalized vegetation index was calculated using the reflectance ratio of near-infrared band to red band. Chlorophyll content index was calculated based on red-edge band features, and plant height was measured using stereoscopic vision technology; The extracted growth parameters are correlated with the corresponding location information to generate a spatialized growth distribution map.

9. A method for cultivating summer vegetables in high-altitude areas according to claim 3, characterized in that, The weighted fusion of the suitability score layers for each environmental parameter to obtain a comprehensive planting suitability distribution map specifically includes: Based on the growth requirements of the target summer vegetable varieties, the weight coefficients of each environmental parameter are determined by the analytic hierarchy process. The environmental parameters include soil moisture, soil temperature, soil pH, air temperature, air humidity, and light intensity. The suitability score layers for each environmental parameter are converted into raster data with the same spatial resolution to ensure that each raster cell corresponds to a consistent geographical location. For each grid cell, the sum of the products of the suitability score of each environmental parameter and the corresponding weight coefficient is calculated to obtain the weighted suitability score of the grid cell. The weighted suitability scores of all grid cells are linearly normalized to map the scores to a range of 0 to 1, generating a comprehensive planting suitability distribution map. To verify the accuracy of the integrated planting suitability distribution map, consistency was checked by comparing the suitability scores of historically successful planting areas.

10. A planting system for summer vegetables in high-altitude areas, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of a method for planting summer vegetables in plateau regions as described in any one of claims 1 to 9.

Images