Agricultural land pollution intelligent acquisition method based on data analysis

Abstract

The invention discloses an agricultural land pollution intelligent collection method based on data analysis, and the method comprises the following steps: 1, collecting land information, analyzing the land information, and obtaining land classification information; 2, selecting a land information acquisition mode according to the land category; 3, collecting environment information of the position of the land, and judging whether environment influence factors exist or not; 4, when there is no environmental influence factor, setting a sensor according to a selected acquisition mode, and acquiring land pollution information; and 5, analyzing the collected land pollution information, and when the land pollution information is abnormal, generating warning information. According to the invention, through the technical means of multi-source data fusion, intelligent analysis, real-time monitoring, dynamic early warning and the like, accurate acquisition and scientific management of agricultural land pollution are realized.

Description

Technical Field

[0001] The present invention relates to the field of acquisition methods, and specifically relates to an intelligent acquisition method for agricultural land pollution based on data analysis. Background Art

[0002] The problem of agricultural land pollution is becoming increasingly serious, posing a threat to the ecological environment and the safety of agricultural products. Traditional land pollution monitoring methods have problems such as low efficiency, high cost, and incomplete data, making it difficult to meet the requirements of modern agriculture for land pollution monitoring.

[0003] In the process of analyzing agricultural land pollution, it is necessary to collect pollution information of agricultural land. In the process of collecting pollution information of agricultural land, an agricultural land pollution collection method will be used.

[0004] Existing agricultural land pollution collection methods have low accuracy in collecting data and low intelligence, which has a certain impact on the use of the safety protection system. Therefore, an intelligent acquisition method for agricultural land pollution based on data analysis is proposed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: how to solve the problem that existing agricultural land pollution collection methods have low accuracy in collecting data and low intelligence, which has a certain impact on the use of the safety protection system, and provides an intelligent acquisition method for agricultural land pollution based on data analysis.

[0006] The present invention solves the above technical problem through the following technical solutions. The present invention includes the following steps:

[0007] Step 1: Collect land information, analyze the land information, and obtain land classification information;

[0008] Step 2: Select a land information collection method according to the land category;

[0009] Step 3: Collect environmental information of the location where the land is located, and determine whether there are environmental impact factors;

[0010] Step 4: When there are no environmental impact factors, set sensors according to the selected collection method to collect land pollution information;

[0011] Step 5: Analyze the collected land pollution information. When the land pollution information is abnormal, a warning message is generated.

[0012] The specific process of collecting land information in Step 1 is as follows:

[0013] Utilize satellite remote sensing images and geographic information systems to obtain the geographical location and topography of the land;

[0014] On-site investigation and sampling: Combining field investigations, soil samples of the land are collected for laboratory analysis to obtain the type of soil;

[0015] Drone aerial photography: Using a drone equipped with a multispectral camera to obtain the vegetation coverage and topographic information of the land;

[0016] The specific process of land classification is as follows:

[0017] Using clustering analysis and decision tree algorithms, classify the land according to geographical location, topography, soil type, terrain and vegetation coverage factors to obtain the first classification and the second classification. When the content of the first classification is the same as that of the second classification, the land classification is obtained. The land classification includes cultivated land, forest land, grassland and construction land;

[0018] When the content of the first classification is different from that of the second classification, data collection is carried out again.

[0019] Furthermore, the specific process of selecting the land information collection method according to the land category is as follows:

[0020] According to the land classification results, when the land is classified as cultivated land, forest land, grassland or construction land, formulate collection rules;

[0021] For cultivated land, collect soil fertility indicators (such as organic matter content, nitrogen, phosphorus and potassium content), soil pollution indicators (such as heavy metal content, organic pollutant concentration) and irrigation water quality;

[0022] For forest land: collect vegetation health indicators (such as chlorophyll content, vegetation coverage rate), soil erosion situation and soil humidity;

[0023] For grassland: collect vegetation coverage rate, soil fertility and soil erosion situation

[0024] For construction land: collect soil heavy metal pollution, organic pollutant concentration and groundwater pollution situation;

[0025] Select sensors, and use soil heavy metal sensors, organic pollutant sensors, pH sensors, soil humidity sensors and vegetation health sensors for information collection;

[0026] According to the land area and pollution risk, deploy sensor nodes in a grid-based sampling method. For example, 5-10 sensor nodes are arranged per hectare of land..

[0027] Plan sampling points. According to the land area and topography, use the stratified random sampling method to determine the sampling points, and at the same time determine the sampling depth according to the monitoring target. For example, 3-5 sampling points are set per hectare of land.

[0028] For example, when monitoring soil fertility, the sampling depth is 0-20 cm; when monitoring heavy metal pollution, the sampling depth is 0-60 cm. Low pollution risk land: Sampling is carried out 1-2 times a year, mainly before and after the crop growing season. High pollution risk land: Sampling is carried out once a quarter, or the sampling frequency is increased as needed.

[0029] Furthermore, the specific process of determining whether there are environmental impact factors is as follows:

[0030] First, collect the environmental information of the location where the land is located, including the rainfall information within the preset time at the location where the land is located, the pollution information of the land irrigation water, and the pollution source information within the preset range of the land;

[0031] When any one of the abnormal rainfall information within the preset time, the abnormal pollution information of the land irrigation water, and the abnormal pollution source information within the preset range of the land appears, it indicates that there are environmental impact factors.

[0032] Furthermore, the determination process of the abnormal rainfall information within the preset time is as follows: The preset time is the time one week before the set collection time point;

[0033] Collect the rainfall information within the preset time, calculate the average value of the rainfall information within the preset time, obtain the rainfall average value. When the rainfall average value is greater than the preset value, it represents the rainfall information within the preset time;

[0034] Then calculate the total amount of rainfall information within the preset time, obtain the total rainfall. When the total rainfall is greater than the preset value, it indicates that the rainfall information is abnormal;

[0035] The determination process of the abnormal pollution information of the land irrigation water is as follows:

[0036] Extract the pollution information of the land irrigation water, including heavy metal content and organic pollutant content;

[0037] When either the heavy metal content or the organic pollutant content is greater than the preset value, it indicates that the pollution information of the land irrigation water is abnormal;

[0038] The determination process of the abnormal pollution source information within the preset range of the land is as follows:

[0039] The pollution source information within the preset range of the land is the number of pollution sources within the preset range around the land where pollution collection needs to be carried out. When the number of pollution sources within the preset range around the land is greater than the preset value, it indicates that the pollution source information within the preset range of the land is abnormal.

[0040] Furthermore, the collection process of the pollution source information within the preset range of the land is as follows:

[0041] Regularly conduct aerial photography of the surrounding area of the land using a drone equipped with a multispectral camera or a thermal imaging camera to identify the number of potential pollution sources (such as factories, farms, landfills, etc.) and obtain the first pollution source quantity;

[0042] Satellite remote sensing monitoring: Use high-resolution satellite images to analyze the distribution quantity of pollution sources within a preset range around the land and obtain the second pollution source quantity;

[0043] Calculate the average value of the first pollution source quantity and the second pollution source quantity, that is, obtain the pollution source information within the preset range of the land.

[0044] Furthermore, when the land pollution information is abnormal, the specific process of generating a warning message is as follows:

[0045] First, conduct data analysis and modeling to calculate the average value, standard deviation, and correlation coefficient of pollution indicators, and analyze the spatio-temporal distribution law of pollution;

[0046] Use algorithms such as support vector machine (SVM), random forest (RF), or neural network to establish a land pollution prediction model to identify pollution anomalies;

[0047] Combine with the geographic information system (GIS) to draw a pollution distribution map and identify pollution hotspots;

[0048] Threshold setting: According to the normal range of pollution indicators, set a threshold. When any one of the collected land pollution information data exceeds the threshold, the system automatically identifies it as abnormal and generates a warning message.

[0049] The present invention has the following advantages compared with the prior art: This intelligent acquisition method for agricultural land pollution based on data analysis has multiple techniques, combining satellite remote sensing, UAV aerial photography, on-site investigation, and sensor networks, etc., to achieve all-round and multi-level acquisition of land information. This multi-source data fusion method can more comprehensively reflect the actual situation of the land, improving the accuracy and reliability of monitoring data. Intelligent data analysis: Through advanced algorithms such as clustering analysis, decision tree algorithms, support vector machines (SVM), random forests (RF), and neural networks, classify land information and predict pollution, and can quickly identify pollution anomalies and hotspots. This intelligent data processing method greatly improves data processing efficiency and analysis accuracy. Using Internet of Things technology, the sensor network can collect land pollution information in real time and transmit the data to the cloud platform through wireless communication technology. This real-time monitoring system can detect pollution problems in a timely manner and avoid further spread of pollution. Automatically identify abnormal data according to preset thresholds and generate warning messages to remind management personnel to take measures. This dynamic monitoring and real-time warning mechanism can effectively reduce pollution risks and improve the response speed of land management. According to the land area and pollution risk, deploy sensors and sampling points in a grid layout and stratified random sampling manner. This scientific layout method can optimize resource allocation, reduce monitoring costs, and ensure the representativeness of monitoring data at the same time. The monitoring strategy can be dynamically adjusted according to land categories and environmental impact factors, such as increasing the monitoring frequency in high-pollution risk areas. This flexible monitoring strategy can improve resource utilization efficiency and avoid resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is the overall flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] The following will describe the embodiments of the present invention in detail. These embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation manners and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments.

[0052] As Figure 1 shown, this embodiment provides a technical solution: An intelligent acquisition method for agricultural land pollution based on data analysis, including the following steps:

[0053] Step 1: Collect land information, analyze the land information, and obtain land classification information;

[0054] Step 2: Select the land information collection method according to the land category;

[0055] Step 3: Collect the environmental information of the location where the land is located and determine whether there are environmental impact factors;

[0056] Step 4: When there are no environmental impact factors, set the sensors according to the selected collection method and collect land pollution information;

[0057] Step 5: Analyze the collected land pollution information. When the land pollution information is abnormal, a warning message is generated.

[0058] The specific process of collecting land information in Step 1 is as follows:

[0059] Use satellite remote sensing images and geographic information systems to obtain the geographical location and topography of the land;

[0060] On-site investigation and sampling: Combine field investigations, collect soil samples of the land, conduct laboratory analyses, and obtain the soil type;

[0061] Drone aerial photography: Use a drone equipped with a multispectral camera to obtain the vegetation coverage and terrain information of the land;

[0062] The specific process of land classification is as follows:

[0063] Use clustering analysis and decision tree algorithms to classify the land according to geographical location, topography, soil type, terrain, and vegetation coverage factors, and obtain the first classification and the second classification. When the first classification content and the second classification content are the same, the land classification is obtained. The land classification includes cultivated land, forest land, grassland, and construction land;

[0064] When the first classification content and the second classification content are different, data collection is restarted.

[0065] Furthermore, the specific process of selecting the land information collection method according to the land category is as follows:

[0066] According to the land classification results, when the land is classified as cultivated land, forest land, grassland, or construction land, formulate collection rules;

[0067] For cultivated land, collect soil fertility indicators (such as organic matter content, nitrogen, phosphorus, and potassium content), soil pollution indicators (such as heavy metal content, organic pollutant concentration), and irrigation water quality;

[0068] For forest land: Collect vegetation health indicators (such as chlorophyll content, vegetation coverage rate), soil erosion conditions, and soil humidity;

[0069] For grassland: Collect vegetation coverage rate, soil fertility, and soil erosion conditions

[0070] For construction land: Collect soil heavy metal pollution, organic pollutant concentration, and groundwater pollution conditions;

[0071] Sensor selection is carried out, and a soil heavy metal sensor, an organic pollutant sensor, a pH sensor, a soil humidity sensor, and a vegetation health sensor are used for information collection;

[0072] According to the land area and pollution risk, sensor nodes are deployed in a grid sampling method. For example, 5-10 sensor nodes are arranged per hectare of land..

[0073] Sampling point planning is carried out. According to the land area and terrain, a stratified random sampling method is used to determine the sampling points, and the sampling depth is determined according to the monitoring objectives. For example, 3-5 sampling points are set per hectare of land.

[0074] For example, when monitoring soil fertility, the sampling depth is 0-20 cm; when monitoring heavy metal pollution, the sampling depth is 0-60 cm. Low pollution risk land: Sampling is carried out 1-2 times a year, mainly before and after the crop growth season. High pollution risk land: Sampling is carried out once a quarter, or the sampling frequency is increased as needed.

[0075] Land information is collected from macro (geographical location, topography, vegetation cover) to micro (soil type, fertility, pollution indicators) in a multi-dimensional manner through a combination of satellite remote sensing, UAV aerial photography, on-site investigation and sampling. This multi-source data fusion method can comprehensively cover the natural and social attributes of the land, avoiding the limitations of a single data source. Data richness: Not only static information of the land (such as topography and soil type) is obtained, but also dynamic information (such as vegetation growth status and soil humidity changes) is obtained through UAVs and sensor networks, providing richer data support for subsequent land classification and monitoring.

[0076] Accuracy and Efficiency Precise Land Classification: Use clustering analysis and decision tree algorithms to classify land, considering multiple factors such as geographical location, topography, soil type, and vegetation cover to ensure the accuracy of land classification. This multi-factor-based classification method can more scientifically identify land types (such as arable land, forest land, grassland, and construction land), providing a reliable basis for formulating subsequent monitoring strategies. Dual Verification Mechanism: Verify the accuracy of classification by obtaining the first classification and the second classification and comparing the results. When the classification results are inconsistent, re-collect data to further ensure the reliability of land classification. 3. Flexibility and Targeting Classification Result-Driven Monitoring Strategies: Develop targeted monitoring rules and sensor configuration plans based on land classification results. For example, focus on monitoring soil fertility and pollution indicators for arable land, and vegetation health and soil erosion for forest land. This classification-driven strategy can flexibly adjust monitoring priorities according to the characteristics and needs of different land types, improving monitoring efficiency. Flexible Deployment of Sensors and Sampling Points: Combine land area and pollution risk, and use a grid-based layout method to deploy sensor nodes, and determine the number of sampling points and sampling depth according to land type and monitoring objectives. This flexible deployment method can optimize resource allocation and ensure the representativeness and accuracy of monitoring data.

[0077] Sustainability and Economy Optimized Monitoring of Low-Pollution-Risk Land: For low-pollution-risk land, adopt a method of regular sampling and analysis to reduce monitoring costs. This optimized monitoring strategy can reduce resource consumption and improve economy while ensuring monitoring effectiveness. Key Monitoring of High-Pollution-Risk Land: For high-pollution-risk land, increase the sampling frequency and sensor layout density to ensure timely detection and treatment of pollution problems, avoid further spread of pollution, and thus reduce subsequent treatment costs.

[0078] Intelligence and Dynamic Adjustment Dynamic Monitoring and Early Warning: Real-time collect land pollution information through a sensor network and identify pollution anomalies in combination with a data analysis model. This dynamic monitoring method can timely detect potential pollution problems and provide early warnings for pollution prevention and control. Intelligent Decision Support: Use a data analysis model and Geographic Information System (GIS) to draw a pollution distribution map, identify pollution hotspots, and provide scientific decision-making suggestions for management personnel. This intelligent support method can improve the scientificity and efficiency of land management.

[0079] The specific process of determining whether there are environmental impact factors is as follows:

[0080] First, collect environmental information at the location of the land, including rainfall information within a preset time at the location of the land, pollution information of the land irrigation water, and pollution source information within the preset range of the land;

[0081] When any one of the abnormal rainfall information within the preset time, abnormal pollution information of land irrigation water, and abnormal pollution source information within the preset range of the land appears, it indicates the existence of environmental impact factors;

[0082] By collecting rainfall information, irrigation water pollution information, and pollution source information, comprehensively evaluate the environmental conditions of the land location from three key dimensions of meteorology, water quality, and pollution sources. This multi-dimensional data collection method can systematically identify potential environmental impact factors and avoid misjudgment caused by missing a single factor.

[0083] Incorporating meteorological conditions, water quality, and the number of pollution sources into the comprehensive evaluation system can comprehensively reflect the complexity and dynamic changes of the environment where the land is located, providing more comprehensive background information for land pollution monitoring. 2. Early warning and risk prevention and control

[0084] By setting preset values, conduct real-time monitoring and anomaly detection of rainfall, irrigation water pollution, and the number of pollution sources. When any one of the indicators exceeds the threshold, it is determined that there are environmental impact factors, and an early warning can be issued in a timely manner. This early warning mechanism helps to detect potential risks in advance, avoid the further spread of pollution, and reduce the long-term damage to the land and the ecosystem.

[0085] Monitoring the rainfall and irrigation water pollution information within the preset time can capture the impact of short-term environmental changes on the land and adjust the monitoring strategy and management measures in a timely manner..

[0086] The determination process of the abnormal rainfall information within the preset time is as follows: The preset time is one week before the set collection time point;

[0087] Collect the rainfall information within the preset time, calculate the average value of the rainfall information within the preset time, obtain the rainfall average value. When the rainfall average value is greater than the preset value, it represents the rainfall information within the preset time;

[0088] Then calculate the total amount of rainfall information within the preset time, obtain the total rainfall. When the total rainfall is greater than the preset value, it indicates that the rainfall information is abnormal;

[0089] The determination process of the abnormal pollution information of land irrigation water is as follows:

[0090] Extract the pollution information of land irrigation water, including heavy metal content and organic pollutant content;

[0091] When either the heavy metal content or the organic pollutant content is greater than the preset value, it indicates that the pollution information of land irrigation water is abnormal;

[0092] The determination process of the abnormal pollution source information within the preset range of the land is as follows:

[0093] The pollution source information within the preset range of the land refers to the number of pollution sources within the preset range around the land that needs to be polluted. When the number of pollution sources within the preset range around the land is greater than the preset value, it indicates that the pollution source information within the preset range of the land is abnormal.

[0094] The process of collecting the pollution source information within the preset range of the land is as follows:

[0095] Regularly conduct aerial photography of the surrounding area of the land by using a drone equipped with a multispectral camera or a thermal imaging camera to identify the number of potential pollution sources (such as factories, farms, landfills, etc.) and obtain the first pollution source number;

[0096] Satellite remote sensing monitoring, using high-resolution satellite images, analyze the distribution quantity of pollution sources within the preset range around the land to obtain the second pollution source number;

[0097] Calculate the average value of the first pollution source number and the second pollution source number, that is, obtain the pollution source information within the preset range of the land.

[0098] The specific process of generating a warning message when the land pollution information is abnormal is as follows:

[0099] First, conduct data analysis and modeling to calculate the average value, standard deviation, and correlation coefficient of pollution indicators, and analyze the temporal and spatial distribution laws of pollution;

[0100] Use algorithms such as support vector machine (SVM), random forest (RF), or neural network to establish a land pollution prediction model to identify pollution anomalies;

[0101] Combine with the geographic information system (GIS) to draw a pollution distribution map and identify pollution hotspots;

[0102] Threshold setting: According to the normal range of pollution indicators, set a threshold. When any data in the collected land pollution information exceeds the threshold, the system automatically identifies it as abnormal and generates a warning message;

[0103] By calculating the average value, standard deviation, and correlation coefficient of pollution indicators to analyze the temporal and spatial distribution laws of pollution, the dynamic changes of pollution can be accurately identified. The prediction model established by combining advanced algorithms such as support vector machine (SVM), random forest (RF), or neural network can effectively identify potential pollution anomalies and avoid misjudgment caused by data fluctuations or randomness.

[0104] When the monitoring data exceeds the preset threshold, the system automatically identifies the anomaly and generates a warning message. This early warning mechanism can timely detect pollution problems, avoid the further spread of pollution, and reduce the long-term damage to the land and ecological environment.

[0105] Visualization of pollution distribution: By combining Geographic Information System (GIS) to draw pollution distribution maps, it can intuitively display the geographical location and distribution range of pollution, helping managers quickly locate pollution hotspots. This visualization tool provides strong support for scientific decision-making and facilitates the formulation of targeted pollution control measures.

[0106] Dynamic monitoring and trend analysis: Through data analysis and modeling, it can monitor the changing trends of pollution indicators in real time, providing managers with dynamic information on the development of pollution. This dynamic monitoring ability helps to plan control measures in advance and avoid the deterioration of pollution problems.

[0107] Precise positioning and resource optimization: By identifying pollution hotspots, managers can concentrate limited resources on the treatment and monitoring of high-risk areas, avoiding waste of resources. This precise positioning and resource optimization strategy can improve treatment efficiency and reduce treatment costs.

[0108] Reduction of unnecessary intervention: Based on data-driven anomaly identification and threshold warning mechanisms, it avoids over-intervention or neglect of problems caused by subjective judgment. This scientific monitoring and warning method can ensure the reasonable allocation and efficient utilization of resources.

[0109] Pollution prevention and control and ecological restoration: Through early warning and precise treatment, it can timely control pollution sources and reduce further damage to soil, water bodies and ecosystems caused by pollutants. This pollution prevention and control mechanism helps to protect biodiversity and promote the restoration and sustainable development of ecosystems.

[0110] Long-term environmental monitoring and management: By combining data analysis and GIS visualization tools, it can establish a long-term environmental monitoring system, providing a scientific basis for the sustainable use of land. This long-term monitoring and management strategy helps to achieve a balance between the rational development of land resources and ecological protection.

[0111] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0112] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples", etc., mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0113] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. An intelligent collection method for agricultural land pollution based on data analysis, characterized in that: The following steps are involved: Step 1: Collect land information, analyze the land information, and obtain land classification information; Step 2: Select the land information collection method based on the land type; Step 3: Collect environmental information about the land location to determine whether there are any environmental impact factors; Step 4: When there are no environmental impact factors, set the sensor according to the selected collection method to collect land pollution information; Step 5: Analyze the collected land pollution information. When the land pollution information is abnormal, generate a warning message.

2. According to claim 1, a method for intelligent collection of agricultural land pollution based on data analysis is characterized in that: The specific process of land information collection in step 1 is as follows: Use satellite remote sensing images and geographic information systems to obtain the geographical location and topography of the land; Field investigation and sampling: Combined with field investigation, soil samples are collected from the land and laboratory analysis is performed to obtain the soil type; Drone aerial photography: Use drones equipped with multispectral cameras to obtain land vegetation coverage and terrain information; The specific process of land classification is as follows: Using cluster analysis and decision tree algorithms, the land is classified according to geographical location, topography, soil type, terrain and vegetation cover factors to obtain the first classification and the second classification. When the first classification content and the second classification content are the same, the land classification is considered as land. The land classification includes cultivated land, forest land, grassland and construction land. When the content of the first category is different from the content of the second category, data collection is performed again.

3. The method for intelligently collecting agricultural land pollution based on data analysis according to claim 1 is characterized by: The specific process of selecting the land information collection method according to the land category is as follows: According to the land classification results, the land is classified as cultivated land, forest land, grassland or construction land, and the collection rules are formulated; For cultivated land, soil fertility indicators, soil pollution indicators and irrigation water quality are collected; Forest land: collect vegetation health indicators, soil erosion and soil moisture; Grassland: Collect vegetation coverage, soil fertility and soil erosion Construction land: collect soil heavy metal pollution, organic pollutant concentration and groundwater pollution; Select sensors and use soil heavy metal sensors, organic pollutant sensors, pH sensors, soil moisture sensors and vegetation health sensors to collect information; Sensor nodes are deployed in a grid-based manner based on the land area and pollution risk. Sampling point planning is carried out, and sampling points are determined by stratified random sampling method according to land area and topography, and sampling depth is determined according to monitoring objectives.

4. The method for intelligently collecting agricultural land pollution based on data analysis according to claim 1 is characterized by: The specific process of determining whether there are environmental influencing factors is as follows: First, collect environmental information about the land location, including rainfall information within a preset time at the land location, pollution information about the land irrigation water, and pollution source information within a preset range of the land; When any of the abnormal rainfall information within the preset time, the abnormal pollution information of the land irrigation water, and the abnormal pollution source information within the preset range of the land appears, it means that there are environmental influencing factors.

5. The method for intelligently collecting agricultural land pollution based on data analysis according to claim 4 is characterized by: The determination process of abnormal rainfall information within the preset time is as follows: the preset time is the time one week forward based on the set collection time point; Collecting rainfall information within a preset time, calculating the mean of the rainfall information within the preset time, and obtaining the mean rainfall value. When the mean rainfall value is greater than the preset value, it indicates the rainfall information within the preset time; Then calculate the sum of the rainfall information within the preset time to obtain the total rainfall. When the total rainfall is greater than the preset value, it means that the rainfall information is abnormal. The process of determining abnormal pollution information of land irrigation water is as follows: Extract pollution information of land irrigation water, including heavy metal content and organic pollutant content; When any one of the heavy metal content and organic pollutant content is greater than the preset value, it means that the pollution information of land irrigation water is abnormal; The process of determining abnormal pollution source information within the preset land area is as follows: The pollution source information within the preset range of the land is the number of pollution sources within the preset range around the land where pollution collection is required. When the number of pollution sources within the preset range around the land is greater than the preset value, it means that the pollution source information within the preset range of the land is abnormal.

6. The method for intelligently collecting agricultural land pollution based on data analysis according to claim 4 is characterized by: The process of collecting pollution source information within the preset scope of the land is as follows: Use drones equipped with multispectral cameras or thermal imaging cameras to regularly take aerial photos of the area surrounding the land to identify the number of potential pollution sources and obtain the number of first pollution sources; Satellite remote sensing monitoring uses high-resolution satellite images to analyze the distribution of pollution sources within a preset range around the land and obtain the number of secondary pollution sources; The average of the number of the first pollution source and the number of the second pollution source is calculated, that is, the pollution source information within the preset range of the land is obtained.

7. The method for intelligently collecting agricultural land pollution based on data analysis according to claim 1 is characterized by: When the land pollution information is abnormal, the specific process of generating warning information is as follows: First, data analysis and modeling are performed to calculate the mean, standard deviation and correlation coefficient of pollution indicators, and to analyze the temporal and spatial distribution of pollution; Use algorithms such as support vector machines, random forests or neural networks to build land pollution prediction models and identify pollution anomalies; Combined with geographic information systems, pollution distribution maps are drawn to identify pollution hot spots; Threshold setting: Set the threshold according to the normal range of pollution indicators. When any data in the collected land pollution information exceeds the threshold, the system automatically identifies it as an abnormality and generates a warning message.

Images