Intelligent crop monitoring system based on spatio-temporal data

Abstract

The invention provides an intelligent crop monitoring system based on spatio-temporal data, and relates to the technical field of crop monitoring, and the system comprises a data collection module which is used for collecting multi-source spatio-temporal data of a target farmland area; the multi-source spatio-temporal data comprises remote sensing images, meteorological data, soil data and farmland sensor data; the data processing module is used for carrying out preprocessing and data fusion on the multi-source spatio-temporal data to obtain a unified spatio-temporal data set; the intelligent analysis module is used for performing intelligent analysis on the growth state, the pest and disease risk and the yield of the crops according to the spatio-temporal data set based on artificial intelligence and spatio-temporal analysis technologies to obtain an analysis result; and the visual display module is used for carrying out visual display on the analysis result. The agricultural production efficiency and the management level are improved, powerful support is provided for sustainable agricultural development, and remarkable economic, social and environmental benefits are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of crop monitoring, and in particular to an intelligent crop monitoring system based on spatiotemporal data. Background Art

[0002] With the growth of global population and the intensification of climate change, agricultural production is facing increasing challenges. Traditional crop monitoring methods mainly rely on manual inspections and simple sensor data collection, which is not only inefficient but also difficult to achieve large-scale and refined crop management. In recent years, the rapid development of remote sensing technology, the Internet of Things (IoT), artificial intelligence (AI) and big data analysis technology has provided new solutions for agricultural intelligence.

[0003] Spatiotemporal data (including remote sensing images, meteorological data, geographic information, etc.) can fully reflect the growth status, environmental conditions and spatial distribution characteristics of crops. However, how to efficiently integrate multi-source spatiotemporal data to build an intelligent crop monitoring system to monitor the growth status of crops in real time, predict yields and provide accurate management suggestions is still a technical problem that needs to be solved. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide an intelligent crop monitoring system based on spatiotemporal data, which not only improves agricultural production efficiency and management level, but also provides strong support for sustainable agricultural development, and has significant economic, social and environmental benefits.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A crop intelligent monitoring system based on spatiotemporal data, comprising:

[0007] A data acquisition module is used to collect multi-source spatiotemporal data of the target farmland area; the multi-source spatiotemporal data includes remote sensing images, meteorological data, soil data and farmland sensor data;

[0008] A data processing module, used for preprocessing and data fusion of the multi-source spatiotemporal data to obtain a unified spatiotemporal data set;

[0009] An intelligent analysis module, for performing intelligent analysis on the growth status, pest and disease risk, and yield of crops according to the spatiotemporal data set based on artificial intelligence and spatiotemporal analysis technology to obtain analysis results;

[0010] The visualization module is used to visualize the analysis results.

[0011] Preferably, the data acquisition module includes:

[0012] A remote sensing image acquisition submodule, used for acquiring the remote sensing image of the target farmland area through satellite remote sensing or unmanned aerial vehicle;

[0013] Meteorological data acquisition submodule: used to obtain the meteorological data through a weather station or online meteorological service; the meteorological data includes temperature, humidity, precipitation and wind speed;

[0014] A soil data collection submodule, used to collect the soil data using a soil sensor; the soil data includes: soil moisture, temperature, pH value and nutrient content;

[0015] The farmland sensor data acquisition submodule is used to monitor the farmland environment in real time based on the deployed Internet of Things sensors to obtain the farmland sensor data.

[0016] Preferably, the data processing module includes:

[0017] An image correction submodule is used to perform cloud interference removal, geometric correction and radiation correction on the remote sensing image to obtain a corrected image;

[0018] An image enhancement submodule, used to perform image enhancement on the corrected image to obtain an enhanced image;

[0019] A data denoising submodule, used for denoising the meteorological data, the soil data and the farmland sensor data to obtain denoised data;

[0020] The data integration submodule is used to integrate the enhanced image and the denoised data using a spatiotemporal data fusion algorithm to obtain the spatiotemporal data set.

[0021] Preferably, the image enhancement submodule includes:

[0022] A filtering unit, used for filtering the corrected image to obtain a filtered image;

[0023] A denoising unit, used for constructing an image enhancement window, and using the image enhancement window to traverse the filtered image for denoising, to obtain a denoised image;

[0024] The enhancement unit is used to perform histogram equalization processing on the denoised image to obtain the enhanced image.

[0025] Preferably, the filtering unit comprises:

[0026] Performing wavelet decomposition on the corrected image at multiple scales to obtain corresponding wavelet coefficients;

[0027] A filtering threshold is constructed according to the size and decomposition scale of the corrected image; the calculation formula of the filtering threshold is: Among them, λ j represents the filtering threshold at the jth decomposition scale, σ j represents the standard deviation of the noise at the jth decomposition scale, N j represents the signal length at the jth decomposition scale, where j represents the decomposition scale;

[0028] The image filtering model is constructed using the filtering threshold; the formula of the image filtering model is: Wherein, sign is the sign function, a is the first preset coefficient, b is the second preset coefficient, ω j,k represents the kth wavelet coefficient at the jth decomposition scale, represents the wavelet coefficients after filtering;

[0029] The corrected image is filtered using the image filtering model to obtain the filtered image.

[0030] Preferably, the formula of the spatiotemporal data fusion algorithm is:

[0031] D fused (x,y,t)=α·D image (x,y,t)+β·D weather (x,y,t)+γ·D soil (x,y,t)+δ·

[0032] D sensor (x,y,t);

[0033] Among them, D fused (x, y, t) is the fused spatiotemporal data value, which represents the comprehensive information at position (x, y) and time t. image (x, y, t) is the corrected remote sensing image data, which represents the image information of the crop growth status at the location (x, y) and time t. weather (x, y, t) is the meteorological data, which indicates the meteorological conditions at the location (x, y) and time t. soil (x, y, t) is soil data, which represents the soil characteristics at location (x, y) and time t. sensor (x, y, t) is the farmland sensor data, which means the real-time monitoring data at the location (x, y) and time t. α, β, γ, and δ are all weight coefficients, corresponding to the influence of remote sensing images, meteorological data, soil data, and farmland sensor data, respectively.

[0034] Preferably, the intelligent analysis module includes:

[0035] A feature extraction submodule is used to extract features from the spatiotemporal data set to obtain key features;

[0036] The model training submodule is used to train the preset initial neural network according to the preset sample key feature set to obtain a trained growth state prediction model;

[0037] The prediction submodule is used to input the key features into the growth status prediction model to obtain the analysis results.

[0038] Preferably, the key features include air temperature, humidity, soil moisture, and NDVI of remote sensing images.

[0039] Preferably, the initial neural network is a CNN-LSTM combined network.

[0040] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0041] The present invention provides an intelligent crop monitoring system based on spatiotemporal data, comprising: a data acquisition module for collecting multi-source spatiotemporal data of a target farmland area; the multi-source spatiotemporal data includes remote sensing images, meteorological data, soil data and farmland sensor data; a data processing module for preprocessing and data fusion of the multi-source spatiotemporal data to obtain a unified spatiotemporal data set; an intelligent analysis module for intelligently analyzing the growth status, pest and disease risk and yield of crops based on the spatiotemporal data set based on artificial intelligence and spatiotemporal analysis technology to obtain analysis results; and a visualization display module for visually displaying the analysis results. The present invention not only improves agricultural production efficiency and management level, but also provides strong support for sustainable agricultural development, and has significant economic, social and environmental benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0043] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The purpose of the present invention is to provide an intelligent crop monitoring system based on spatiotemporal data, which not only improves agricultural production efficiency and management level, but also provides strong support for sustainable agricultural development, with significant economic, social and environmental benefits.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides an intelligent crop monitoring system based on spatiotemporal data, comprising:

[0048] A data acquisition module is used to collect multi-source spatiotemporal data of the target farmland area; the multi-source spatiotemporal data includes remote sensing images, meteorological data, soil data and farmland sensor data;

[0049] A data processing module, used for preprocessing and data fusion of the multi-source spatiotemporal data to obtain a unified spatiotemporal data set;

[0050] An intelligent analysis module, for performing intelligent analysis on the growth status, pest and disease risk, and yield of crops according to the spatiotemporal data set based on artificial intelligence and spatiotemporal analysis technology to obtain analysis results;

[0051] The visualization module is used to visualize the analysis results.

[0052] Preferably, the data acquisition module includes:

[0053] A remote sensing image acquisition submodule, used for acquiring the remote sensing image of the target farmland area through satellite remote sensing or unmanned aerial vehicle;

[0054] Meteorological data acquisition submodule: used to obtain the meteorological data through a weather station or online meteorological service; the meteorological data includes temperature, humidity, precipitation and wind speed;

[0055] A soil data collection submodule, used to collect the soil data using a soil sensor; the soil data includes: soil moisture, temperature, pH value and nutrient content;

[0056] The farmland sensor data acquisition submodule is used to monitor the farmland environment in real time based on the deployed Internet of Things sensors to obtain the farmland sensor data.

[0057] Specifically, the remote sensing image acquisition submodule obtains remote sensing images of the target farmland area through satellite remote sensing or drone technology. Satellite remote sensing uses sensors carried by satellites to regularly photograph the earth's surface and provide a wide range of farmland image data, which is suitable for regional monitoring and analysis. Drones can fly at low altitudes at specific times and locations to obtain high-resolution images, which can capture more detailed crop growth conditions and environmental changes. This submodule needs to be equipped with high-performance image sensors and data transmission systems to ensure the clarity and real-time nature of the images. At the same time, combined with image processing technology, the acquired images are preliminarily processed, such as cloud removal, geometric correction, and radiation correction, to improve data quality.

[0058] The meteorological data acquisition submodule obtains meteorological data of the target farmland area through meteorological stations or online meteorological services. The meteorological station can monitor meteorological parameters such as temperature, humidity, precipitation and wind speed in real time, and transmit the data to the central database. Online meteorological services provide historical and real-time meteorological data, which can supplement the shortcomings of meteorological stations. This submodule needs to be effectively connected with meteorological equipment and service platforms to ensure the accuracy and timeliness of data. In addition, the frequency of meteorological data collection should be adjusted according to the characteristics of crop growth cycle and climate change to better support subsequent intelligent analysis.

[0059] The soil data collection submodule uses soil sensors to monitor various soil indicators in real time, including soil moisture, temperature, pH value and nutrient content. Soil sensors can be deployed in different locations in the farmland to collect data regularly or in real time, and transmit the data to the data processing module via a wireless network. To ensure the accuracy of the data, the sensors need to be calibrated and regularly maintained and inspected. In addition, the collection of soil data should take into account different soil types and crop requirements to ensure that the acquired data can reflect the actual condition of the soil and provide a scientific basis for subsequent agricultural management.

[0060] The farmland sensor data acquisition submodule monitors the farmland environment in real time based on the deployed IoT sensors and obtains farmland sensor data. These sensors can monitor a variety of environmental factors, such as soil moisture, temperature, light intensity, and air quality. Through IoT technology, sensors can transmit the collected data to the cloud or local server in real time for centralized management and analysis. This submodule needs to ensure the stability of sensors and the reliability of data transmission, while adopting data encryption and security measures to protect the privacy and integrity of data. By comprehensively analyzing farmland sensor data, the system can identify environmental changes in a timely manner, help farmers respond quickly, and optimize agricultural management.

[0061] Preferably, the data processing module includes:

[0062] An image correction submodule is used to perform cloud interference removal, geometric correction and radiation correction on the remote sensing image to obtain a corrected image;

[0063] An image enhancement submodule, used to perform image enhancement on the corrected image to obtain an enhanced image;

[0064] A data denoising submodule, used for denoising the meteorological data, the soil data and the farmland sensor data to obtain denoised data;

[0065] The data integration submodule is used to integrate the enhanced image and the denoised data using a spatiotemporal data fusion algorithm to obtain the spatiotemporal data set.

[0066] Specifically, the first step of the data processing module is the image correction submodule, whose main function is to remove cloud interference, geometric correction and radiation correction of the collected remote sensing images. First, cloud interference removal can be achieved through image processing algorithms (such as multispectral image analysis) to reduce the impact of clouds on image quality. Next, geometric correction is to align the image with the geographic coordinate system through geographic information system (GIS) technology to ensure the accuracy of the spatial position of the image. Radiation correction is to eliminate the response deviation of the sensor by adjusting the brightness and contrast of the image. After the correction is completed, the image enhancement submodule performs wavelet decomposition on the corrected image through the filtering unit to obtain the wavelet coefficients, and constructs the filtering threshold according to the noise standard deviation and signal length. Using this threshold, an image filtering model is constructed to filter the corrected image, and finally a filtered image is obtained.

[0067] After the image enhancement is completed, the data denoising submodule denoises the meteorological data, soil data and farmland sensor data. By applying statistical methods (such as moving average or median filtering), outliers and noise in the data are removed to ensure the accuracy of subsequent analysis. The data integration submodule uses a spatiotemporal data fusion algorithm to integrate the enhanced remote sensing image with the denoised meteorological, soil and sensor data. The algorithm comprehensively considers the impact of various types of data on the growth status of crops based on the weight coefficients of different data sources. The fused spatiotemporal data set can provide comprehensive farmland information and provide basic data support for subsequent intelligent analysis.

[0068] The implementation of the intelligent analysis module is divided into three submodules: feature extraction, model training, and prediction. First, the feature extraction submodule extracts key features from the spatiotemporal data set, including temperature, humidity, soil moisture, and NDVI (normalized difference vegetation index) of remote sensing images. These features can effectively reflect the growth status of crops. Next, the model training submodule uses a preset set of sample key features to train the initial neural network (such as a CNN-LSTM combined network) to learn the relationship between different features and the growth status of crops. After training, the prediction submodule inputs the new key features into the trained growth status prediction model to generate analysis results, helping farmers or agricultural managers to understand the growth status and potential risks of crops in real time, so as to make scientific decisions.

[0069] Preferably, the image enhancement submodule includes:

[0070] A filtering unit, used for filtering the corrected image to obtain a filtered image;

[0071] A denoising unit, used for constructing an image enhancement window, and using the image enhancement window to traverse the filtered image for denoising, to obtain a denoised image;

[0072] The enhancement unit is used to perform histogram equalization processing on the denoised image to obtain the enhanced image.

[0073] Preferably, the filtering unit comprises:

[0074] Performing wavelet decomposition on the corrected image at multiple scales to obtain corresponding wavelet coefficients;

[0075] A filtering threshold is constructed according to the size and decomposition scale of the corrected image; the calculation formula of the filtering threshold is: Among them, λ j represents the filtering threshold at the jth decomposition scale, σ j represents the standard deviation of the noise at the jth decomposition scale, N j represents the signal length at the jth decomposition scale, where j represents the decomposition scale;

[0076] The image filtering model is constructed using the filtering threshold; the formula of the image filtering model is: Wherein, sign is the sign function, a is the first preset coefficient, b is the second preset coefficient, ω j,k represents the kth wavelet coefficient at the jth decomposition scale, represents the wavelet coefficients after filtering;

[0077] The corrected image is filtered using the image filtering model to obtain the filtered image.

[0078] Preferably, the formula of the spatiotemporal data fusion algorithm is:

[0079] D fused (x,y,t)=α·D image (x,y,t)+β·D weather (x,y,t)+γ·D soil (x,y,t)+δ·

[0080] D sensor (x,y,t);

[0081] Among them, D fused (x, y, t) is the fused spatiotemporal data value, which represents the comprehensive information at position (x, y) and time t. image (x, y, t) is the corrected remote sensing image data, which represents the image information of the crop growth status at the location (x, y) and time t. weather (x, y, t) is the meteorological data, which indicates the meteorological conditions at the location (x, y) and time t. soil (x, y, t) is soil data, which represents the soil characteristics at location (x, y) and time t. sensor (x, y, t) is the farmland sensor data, which means the real-time monitoring data at the location (x, y) and time t. α, β, γ, and δ are all weight coefficients, corresponding to the influence of remote sensing images, meteorological data, soil data, and farmland sensor data, respectively.

[0082] Specifically, the first step in determining each weight coefficient of the present embodiment is to perform data correlation analysis. The correlation between remote sensing images, meteorological data, soil data, and farmland sensor data and crop growth status is evaluated by statistical analysis methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.). Specifically, each data type can be compared with the growth indicators of crops (such as yield, health status, etc.) to calculate the correlation coefficient. The larger the absolute value of the correlation coefficient, the more significant the impact of the data type on the growth status of crops. In this way, the relative importance of each data source can be preliminarily determined, providing a basis for subsequent weight allocation.

[0083] After completing the correlation analysis, the present embodiment further optimizes the determination of the weight coefficient using a machine learning model. Select a suitable regression model (such as linear regression, random forest regression, etc.), use remote sensing images, meteorological data, soil data and farmland sensor data as input features, and train the crop growth status as the target variable. Through model training, the model will automatically learn the degree of influence of each feature on the target variable and generate corresponding feature importance scores. These scores can be directly used as a preliminary estimate of the weight coefficient. In order to improve the generalization ability of the model, methods such as cross-validation can be used to ensure the stability and reliability of the weight coefficient. After obtaining the preliminary weight coefficient, it is necessary to normalize it to ensure that the sum of all weight coefficients is 1. This can be achieved by dividing each weight coefficient by the sum of all weight coefficients. The normalized weight coefficient can more intuitively reflect the relative impact of each data source on the crop growth status. In addition, the weight coefficient can be fine-tuned according to expert knowledge or field experience. For example, certain specific meteorological conditions (such as extreme weather) may have a greater impact on crop growth, and the weight of the relevant data can be appropriately increased at this time. In this way, the final weight coefficient is not only based on the data analysis results, but also combined with the needs of actual agricultural management to ensure its scientificity and practicality.

[0084] Preferably, the intelligent analysis module includes:

[0085] A feature extraction submodule is used to extract features from the spatiotemporal data set to obtain key features;

[0086] The model training submodule is used to train the preset initial neural network according to the preset sample key feature set to obtain a trained growth state prediction model;

[0087] The prediction submodule is used to input the key features into the growth status prediction model to obtain the analysis results.

[0088] Preferably, the key features include air temperature, humidity, soil moisture, and NDVI of remote sensing images.

[0089] Preferably, the initial neural network is a CNN-LSTM combined network.

[0090] Specifically, the feature extraction submodule of this embodiment first conducts an in-depth analysis of the spatiotemporal data set to identify key features related to the growth status of crops. The process includes extracting vegetation indices (such as NDVI) from remote sensing images, extracting climate factors such as temperature, humidity, and precipitation from meteorological data, and extracting indicators such as soil moisture, pH value, and nutrient content from soil data. Image processing techniques (such as edge detection, regional growth, etc.) and statistical analysis methods (such as principal component analysis, feature selection algorithms, etc.) can be used to extract and filter features. Ultimately, the feature extraction submodule will generate a feature set containing all key features to provide basic data for subsequent model training.

[0091] Further, the model training submodule of the present embodiment involves selecting a suitable neural network architecture and training. According to the result of feature extraction, an initial neural network model is constructed, such as a CNN-LSTM combined network, so as to process spatial features and time series data simultaneously. During the training process, a preset sample key feature set is used as input, and the growth status of crops (such as health, damage, pests and diseases, etc.) is used as the target output. Through the back propagation algorithm and optimizer (such as Adam or SGD), the weights and biases in the network are adjusted to minimize the loss function between the predicted results and the actual results. In order to improve the generalization ability of the model, cross-validation, early stopping and other techniques can be used to ensure that the model performs well on unseen data.

[0092] After the model training is completed, the prediction submodule will use the trained growth status prediction model for actual prediction. First, the new key features are input into the trained neural network model, and the model will calculate the growth status prediction results of the crops based on the input features. In order to ensure the accuracy of the prediction, the input features can be standardized to make them consistent with the feature distribution during training. In addition, the prediction submodule can also set thresholds to convert continuous prediction results into discrete growth status categories (such as healthy, slightly damaged, severely damaged, etc.) to facilitate subsequent decision support.

[0093] Finally, in order to continuously improve the accuracy and reliability of predictions, a result feedback mechanism is essential. The prediction submodule compares the analysis results with the actual crop growth status and collects feedback data. This feedback data can be used to further optimize the model training process and adjust feature selection and model parameters. The training data set is updated regularly and the model is retrained to adapt to new environmental changes and agricultural practices. In addition, combined with expert knowledge and field experience, the model can be fine-tuned to ensure its effectiveness and adaptability in practical applications. Through this continuous feedback and optimization mechanism, the intelligent analysis module can continuously improve its prediction ability and provide more accurate decision support for agricultural management.

[0094] In addition, the visualization display module of this embodiment first needs to prepare and process the analysis results. This process includes extracting the prediction results and related key feature data from the intelligent analysis module, such as the growth status of crops, the risk level of pests and diseases, and yield forecasts. In order to ensure the clarity and intuitiveness of the visualization effect, the data needs to be sorted and formatted, usually stored in a table or JSON format. In addition, for different types of data (such as time series data, spatial data, etc.), necessary aggregation and statistical analysis can be performed to provide more insightful information during visualization. For example, the average growth status or pest and disease risk in different regions can be calculated, and the corresponding geographic information data can be prepared for subsequent map visualization.

[0095] After the data preparation is completed, the key to realizing the visualization display module is to select appropriate visualization tools and technologies in this embodiment. Interactive charts and maps can be created using open source visualization libraries (such as D3.js, Chart.js, Plotly, etc.) or GIS tools (such as Leaflet, ArcGIS, etc.). For time series data, line graphs or bar graphs can be used to show the changing trend of crop growth status; for spatial data, heat maps or graded symbol maps can be used to show the pest and disease risks and growth status in different regions. In addition, combined with the user interface design principles, ensure that the visualization interface is friendly and easy to operate, allowing users to select different time periods, regions or data types through interactive controls (such as sliders, drop-down menus, etc.) for in-depth analysis.

[0096] Finally, after completing the visualization display, this embodiment implements the display of the results and the user interaction function. By embedding the visualization results into the user interface, the user can intuitively view the analysis results. A dashboard can be designed to centrally display multiple visualization components, such as charts, maps, and data tables, to provide comprehensive agricultural monitoring information. At the same time, the user interaction experience is enhanced, allowing the user to click on a specific area on a chart or map to obtain detailed information and data prompts. For example, when a user clicks on a heat map of an area, a pop-up window can be displayed to show the specific growth status, meteorological data, and soil information of the area. In this way, the visualization display module not only provides an intuitive presentation of the data, but also enhances the user's sense of participation and decision-making ability, helping farmers and agricultural managers to better understand and use the analysis results.

[0097] The beneficial effects of the present invention are as follows:

[0098] (1) The present invention monitors and analyzes the growth status of crops in real time, so that farmers can obtain key information on crop growth in a timely manner, thereby optimizing management decisions and improving agricultural production efficiency.

[0099] (2) The present invention can integrate multi-source spatiotemporal data to provide accurate farmland management suggestions, helping farmers to accurately apply fertilizers, irrigate, and control pests and diseases according to specific meteorological, soil, and crop conditions, thereby reducing resource waste.

[0100] (3) Through the intelligent analysis module, the system of the present invention can evaluate the risk of pests and diseases in real time, issue early warnings, and help farmers take preventive measures to reduce the impact of pests and diseases on crops and reduce economic losses.

[0101] (4) The data analysis results provided by the present invention provide a scientific basis for agricultural managers and farmers, helping them make more reasonable decisions and improving the scientificity and effectiveness of agricultural management.

[0102] (5) The visualization display module of the present invention presents complex data analysis results in an intuitive manner, helping users to quickly understand the growth status and potential risks of crops and enhance the user's decision-making ability.

[0103] (6) The present invention reduces the use of chemical fertilizers and water resources through precise fertilization and irrigation, promotes sustainable agricultural development, and reduces negative impacts on the environment.

[0104] (7) The present invention is able to continuously accumulate and analyze spatiotemporal data, so that farmers can gain deeper insights into agricultural management over time, promoting continuous improvement and optimization of agricultural production.

[0105] (8) By using the intelligent monitoring system, the present invention enables farmers to improve their understanding and application of modern agricultural technology and enhance their technical literacy in agricultural production.

[0106] (9) The implementation of the present invention has promoted the innovation and development of agricultural science and technology, promoted the application of new technologies and methods in agriculture, and promoted the process of agricultural modernization.

[0107] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0108] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An intelligent crop monitoring system based on spatiotemporal data, characterized in that: include: A data acquisition module is used to collect multi-source spatiotemporal data of the target farmland area; the multi-source spatiotemporal data includes remote sensing images, meteorological data, soil data and farmland sensor data; A data processing module, used for preprocessing and data fusion of the multi-source spatiotemporal data to obtain a unified spatiotemporal data set; An intelligent analysis module, for performing intelligent analysis on the growth status, pest and disease risk, and yield of crops according to the spatiotemporal data set based on artificial intelligence and spatiotemporal analysis technology to obtain analysis results; The visualization module is used to visualize the analysis results.

2. The intelligent crop monitoring system based on spatiotemporal data according to claim 1, characterized in that: The data acquisition module includes: A remote sensing image acquisition submodule, used for acquiring the remote sensing image of the target farmland area through satellite remote sensing or unmanned aerial vehicle; Meteorological data acquisition submodule: used to obtain the meteorological data through a weather station or online meteorological service; the meteorological data includes temperature, humidity, precipitation and wind speed; A soil data collection submodule, used to collect the soil data using a soil sensor; the soil data includes: soil moisture, temperature, pH value and nutrient content; The farmland sensor data acquisition submodule is used to monitor the farmland environment in real time based on the deployed Internet of Things sensors to obtain the farmland sensor data.

3. The intelligent crop monitoring system based on spatiotemporal data according to claim 1 is characterized in that: The data processing module comprises: An image correction submodule is used to perform cloud interference removal, geometric correction and radiation correction on the remote sensing image to obtain a corrected image; An image enhancement submodule, used to perform image enhancement on the corrected image to obtain an enhanced image; A data denoising submodule, used for denoising the meteorological data, the soil data and the farmland sensor data to obtain denoised data; The data integration submodule is used to integrate the enhanced image and the denoised data using a spatiotemporal data fusion algorithm to obtain the spatiotemporal data set.

4. The intelligent crop monitoring system based on spatiotemporal data according to claim 3 is characterized in that: The image enhancement submodule comprises: A filtering unit, used for filtering the corrected image to obtain a filtered image; A denoising unit, used for constructing an image enhancement window, and using the image enhancement window to traverse the filtered image for denoising, to obtain a denoised image; The enhancement unit is used to perform histogram equalization processing on the denoised image to obtain the enhanced image.

5. The intelligent crop monitoring system based on spatiotemporal data according to claim 4 is characterized in that: The filtering unit comprises: Performing wavelet decomposition on the corrected image at multiple scales to obtain corresponding wavelet coefficients; A filtering threshold is constructed according to the size and decomposition scale of the corrected image; the calculation formula of the filtering threshold is: Among them, λ j represents the filtering threshold at the jth decomposition scale, σ j represents the standard deviation of the noise at the jth decomposition scale, N j represents the signal length at the jth decomposition scale, where j represents the decomposition scale; The image filtering model is constructed using the filtering threshold; the formula of the image filtering model is: Wherein, sign is the sign function, a is the first preset coefficient, b is the second preset coefficient, ω j,k represents the kth wavelet coefficient at the jth decomposition scale, represents the wavelet coefficients after filtering; The corrected image is filtered using the image filtering model to obtain the filtered image.

6. The intelligent crop monitoring system based on spatiotemporal data according to claim 3 is characterized in that: The formula of the spatiotemporal data fusion algorithm is: D fused (x,y,t)=α·D image (x,y,t)+β·D weather (x,y,t)+γ·D soil (x,y,t)+δ· D sensor (x,y,t); Among them, D fused (x, y, t) is the fused spatiotemporal data value, which represents the comprehensive information at position (x, y) and time t. image (x, y, t) is the corrected remote sensing image data, which represents the image information of the crop growth status at the location (x, y) and time t. weather (x, y, t) is the meteorological data, which indicates the meteorological conditions at the location (x, y) and time t. soil (x, y, t) is soil data, which represents the soil characteristics at location (x, y) and time t. sensor (x, y, t) is the farmland sensor data, which means the real-time monitoring data at the location (x, y) and time t. α, β, γ, and δ are all weight coefficients, corresponding to the influence of remote sensing images, meteorological data, soil data, and farmland sensor data, respectively.

7. The intelligent crop monitoring system based on spatiotemporal data according to claim 1, characterized in that: The intelligent analysis module includes: A feature extraction submodule is used to extract features from the spatiotemporal data set to obtain key features; The model training submodule is used to train the preset initial neural network according to the preset sample key feature set to obtain a trained growth state prediction model; The prediction submodule is used to input the key features into the growth status prediction model to obtain the analysis results.

8. The intelligent crop monitoring system based on spatiotemporal data according to claim 7, characterized in that: The key features include air temperature, humidity, soil moisture, and NDVI of remote sensing images.

9. The intelligent crop monitoring system based on spatiotemporal data according to claim 7, characterized in that: The initial neural network is a CNN-LSTM combined network.

Images