An ecological environment monitoring feedback system

Abstract

The present invention provides an ecological environment monitoring feedback system, comprising a dynamic data acquisition module, which respectively acquires real-time data collected by detection parts in a water area and a region as dynamic analysis factors; a similar data comparison module, which compares the dynamic analysis factors corresponding to the water area with the dynamic analysis factors corresponding to the region, and generates a dynamic factor comparison table based on the comparison results; a marking data analysis module, which selects abnormal factors as marking factors in the dynamic factor comparison table, obtains abnormal diffusion rates through diffusion calculation based on the marking factors in the water area and the marking factors in the region within a certain period of time, and then determines the diffusion direction based on the abnormal diffusion rate, the diffusion direction including diffusion from the water area to the region or diffusion from the region to the water area; and an early warning decision output module, which indexes the abnormal processing scheme under the corresponding abnormal type in the decision database according to the abnormal diffusion rate and the diffusion direction. The present invention has the advantages of real-time monitoring of environmental dynamic factors and improved response speed and efficiency of environmental protection.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental ecological monitoring, and more particularly to an ecological environment monitoring feedback system. Background Art

[0002] As technology evolves, remote sensing technology and big data-based monitoring are used to monitor and provide feedback on the environment's ecology. However, remote sensing technology may have a low spatial resolution, making it difficult to accurately identify specific objects on the ground. Cloud cover, changes in lighting conditions, and other factors may affect the acquisition and quality of remote sensing data. Big data monitoring technology requires monitoring stations or monitoring personnel to manually input various measured data as records, and to make inferences or analyses based on the recorded data, which cannot meet the needs of real-time monitoring. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide an ecological environment monitoring feedback system.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] An ecological environment monitoring feedback system, including

[0006] Dynamic data acquisition module, which obtains real-time data collected from water areas and test pieces in the region as dynamic analysis factors;

[0007] The module for comparing similar data compares the dynamic analysis factors corresponding to the water area with those corresponding to the region, or compares them with similar standard thresholds, and generates a dynamic factor comparison table based on the comparison results;

[0008] The marker data analysis module selects an abnormal factor as a marker factor from the dynamic factor comparison table, calculates an abnormal diffusion rate based on the marker factor in the water area and the marker factor in the region at a certain time, and then determines the diffusion direction based on the abnormal diffusion rate, wherein the diffusion direction includes diffusion from the water area to the region or diffusion from the region to the water area;

[0009] The early warning decision output module indexes the exception handling solution corresponding to the exception type in the decision database according to the exception diffusion rate and diffusion direction.

[0010] Furthermore, it also includes a model building module and a model updating module.

[0011] The model building module obtains historical data and performs training, testing and verification in a certain proportion to obtain a wind factor interference model;

[0012] The model update module selects detection points with the same direction as the diffusion direction based on the measured or recorded real-time wind direction and defines them as interference points, defines the remaining detection points as analysis points, inputs the measured or recorded real-time wind speed into the wind factor interference model to obtain a theoretical abnormal diffusion rate, compares the theoretical diffusion rate with the actual abnormal diffusion rate of the interference point, and selects whether to update the wind factor interference model based on the comparison result.

[0013] Furthermore, when the diffusion direction is from water area to land area, the multiple detection points include eight azimuth points located around the water area.

[0014] Furthermore, the dynamic data acquisition module acquires video images of the region collected by the drone as vegetation cover images;

[0015] It also includes an abnormal factor feedback module, which analyzes the vegetation coverage rate of each detection point through color analysis based on the vegetation coverage image, compares the vegetation coverage rate of each detection point with the coverage threshold, and judges whether there is a vegetation missing abnormality at the corresponding detection point based on the comparison result. If so, the detection point with the vegetation missing abnormality is defined as a missing point, and the remaining detection points are defined as normal points. The marking factor of the missing point is compared with the marking factor of the normal point. If the marking factor of the missing point is greater than the marking factor of the normal point, the instruction corresponding to the vegetation missing and abnormal factor is output; otherwise, the instruction for vegetation missing to be determined is output.

[0016] Furthermore, the abnormal factor feedback module divides the vegetation coverage area and the vegetation missing area in the region according to the vegetation coverage image through the semantic segmentation model, and then uses the contour detection algorithm to perform contour planning on the vegetation coverage area and the vegetation missing area to obtain a regional contour map, and maps each detection point to the regional contour map, and records the number and type of marking factors falling into the vegetation missing area. According to the number of the marking factors, if it is greater than or equal to 2, the interference factor screening instruction is output; otherwise, the type of the output marking factor is an abnormal factor of vegetation missing.

[0017] Furthermore, the detection part includes an insertion rod, which includes at least two rod sections. The two adjacent rod sections can rotate and extend. The outer surface of each rod section is provided with an opening that passes through the rod section. A through channel is formed in the opening, and a number of detection sensors are provided in the through channel.

[0018] Furthermore, the insertion rod is a water area insertion rod, and the water area insertion rod includes three rod sections, namely the first rod section, the second rod section and the third rod section. The first rod section has a mud discharge hole at one end away from the second rod section. The mud discharge hole is located on the outer surface of the first rod section. The mud discharge hole is communicated with a through channel. A soil detection sensor is provided in the through channel of the first rod section, a water quality detection sensor is provided in the through channel of the second rod section, and a gas detection sensor is provided in the through channel of the third rod section.

[0019] Furthermore, the insertion rod is a regional insertion rod, and the regional insertion rod includes two rod sections, and the end of any rod section away from the other rod section is a sharp end.

[0020] Furthermore, it also includes a first adjustment module, which obtains the real-time wind direction measured or recorded externally as the detected wind direction, and controls the rotation of the rod with the gas sensor so that the through channel of the rod is aligned with the wind direction.

[0021] Furthermore, it also includes a second regulating module, which controls the first section of the rod to extend outward from the second section of the rod when it is necessary to obtain data on sediment in the water area at the next time point, so that the sediment at the bottom of the water area enters through the through channel.

[0022] The beneficial effects of the present invention are as follows: the present invention can collect various environmental data in water areas and regions in real time through the dynamic data acquisition module, thereby ensuring that any abnormal changes in the ecological environment can be discovered in a timely manner. This real-time monitoring capability greatly improves the response speed and efficiency of environmental protection. The real-time data can be compared with the standard threshold through the similar data comparison module, and the abnormal factors can be accurately identified, which not only helps to quickly locate the problem, but also provides strong data support for subsequent decision-making; at the same time, through the diffusion calculation analysis of the diffusion rate and direction of the abnormal factors, the severity and scope of the problem can be further understood. The early warning decision output module is used to index the corresponding abnormal handling plan in the decision database according to the abnormal diffusion rate and diffusion direction. This data-based and scientific decision-making method greatly improves the pertinence and effectiveness of ecological environment protection and governance. It can provide customized solutions according to specific circumstances, avoiding blind response and ineffective investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is the overall flow chart of the present invention;

[0024] Figure 2 It is a comparison diagram of vegetation loss and abnormal factors in the present invention;

[0025] Figure 3 It is a structural diagram of the water area insertion rod in the present invention.

[0026] Additional markings: 1. First section of the pole; 2. Second section of the pole; 3. Third section of the pole; 4. Mud discharge hole; 5. Through channel. DETAILED DESCRIPTION

[0027] The present invention will be described in further detail below with reference to the accompanying drawings and embodiments. Identical components are denoted by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "upper," and "lower" used in the following description refer to directions in the accompanying drawings, and the terms "bottom," "top," "inner," and "outer" refer to directions toward or away from the geometric center of a particular component, respectively.

[0028] Currently, the means of monitoring and feedback of the ecological environment include remote sensing technology and big data monitoring. However, the spatial resolution of remote sensing technology may be low, making it difficult to accurately identify specific objects on the ground. Factors such as cloud cover and changes in lighting conditions may affect the acquisition and quality of remote sensing data. For big data monitoring technology, it is necessary to manually input various measured data as records through monitoring stations or monitoring personnel, and to make inferences or analyses based on the recorded data, which cannot meet the needs of real-time monitoring. Therefore, the present invention designs an ecological environment monitoring and feedback system, such as Figure 1 Shown, including

[0029] The dynamic data acquisition module obtains the real-time data collected by the detection parts in the water area and the region as dynamic analysis factors. The water area includes lakes, rivers, etc., and the region includes the land around lakes and rivers. The dynamic data acquisition module is responsible for obtaining environmental data in real time from various sensors and monitoring stations. The environmental data includes dissolved oxygen content, pH value, heavy metal content, etc. in water quality, temperature, humidity, dust particles, polluted gases, etc. in meteorology, and soil pollutants, pH value, soil fertility, etc. in soil. All the above environmental data are called dynamic analysis factors.

[0030] The similar data comparison module compares the dynamic analysis factors corresponding to the water area with the dynamic analysis factors corresponding to the region, or compares them with similar standard thresholds respectively, and generates a dynamic factor comparison table based on the comparison results. The dynamic analysis factors are classified according to the type of data, and then compared item by item. A reasonable threshold is set for each dynamic analysis factor, and the actual data is compared with the threshold. The compared records are output into a dynamic factor comparison table.

[0031] The tag data analysis module selects abnormal factors from the dynamic factor comparison table as tag factors. At a certain time, the abnormal diffusion rate is calculated based on the tag factors in the water area and the tag factors in the region. The diffusion direction is then determined based on the abnormal diffusion rate. The diffusion direction includes diffusion from the water area to the region or from the region to the water area. Specifically, data exceeding the threshold in the dynamic factor comparison table is considered abnormal. Of course, if pollution occurs, there will be at least one type of abnormality. All abnormal dynamic analysis factors are then summarized in the template to facilitate subsequent processing.

[0032] Suppose that there are excessive heavy metals in the water area. At time point A, the heavy metal content in the water area collected by the sensor of the detection part is 30%, and the heavy metal content in the area collected by the sensor of the detection part is 4%. The labeling factor is the heavy metal content. After a certain period of time, at time point B, the heavy metal content in the water area collected by the sensor of the detection part is 26%, and the heavy metal content in the area is 12%. The diffusion rate of the heavy metal can be obtained based on the two contents, and the data can be used to determine that the diffusion direction this time is from the water area to the area.

[0033] The early warning decision output module indexes the exception handling plan for the corresponding exception type in the decision database based on the exception diffusion rate and diffusion direction. Prior to this, a database containing various exception types, diffusion conditions and corresponding handling plans is established in advance, and the early warning information (including exception type, diffusion condition, handling suggestions, etc.) is sent to relevant personnel or institutions via email, SMS, system notification, etc.

[0034] Through the dynamic data acquisition module, the present invention can collect various environmental data in water areas and regions in real time, thereby ensuring that any abnormal changes in the ecological environment can be discovered in a timely manner. This real-time monitoring capability greatly improves the response speed and efficiency of environmental protection. The real-time data can be compared with the standard threshold through the similar data comparison module, and the abnormal factors can be accurately identified, which not only helps to quickly locate the problem, but also provides strong data support for subsequent decision-making; at the same time, through the diffusion calculation and analysis of the diffusion rate and direction of the abnormal factors, the severity and scope of the problem can be further understood. The early warning decision output module is used to index the corresponding abnormal handling plan in the decision database according to the abnormal diffusion rate and diffusion direction. This data-based and scientific decision-making method greatly improves the pertinence and effectiveness of ecological environment protection and governance. It can provide customized solutions according to specific circumstances, avoiding blind response and ineffective investment.

[0035] In addition, if Figure 1 As shown, it also includes a model building module and a model updating module.

[0036] The model building module obtains historical data in a certain proportion for training, testing and verification to obtain a wind factor interference model. Specifically, first, a large amount of historical data needs to be collected. The historical data contains information about the relationship between wind factors (such as wind speed and wind direction) and the required data (pollutant concentration and diffusion rate). Secondly, the data obtained above must be processed to ensure data quality and consistency, and then a suitable model (neural network) is selected for training. For example, the data above can be divided into 70% training set, 20% test set and 10% validation set. Finally, the accuracy of the model is evaluated through indicators to ensure that the wind factor interference model can accurately predict the impact of wind factors on diffusion rate.

[0037] The model update module includes real-time data acquisition: real-time wind direction measured or recorded externally; detection point classification: since wind in the same direction has a significant impact on the diffusion rate of the marker factor, the detection points in the same direction as the diffusion direction are screened out according to the real-time wind direction and defined as interference points, and the remaining detection points are defined as analysis points; theoretical abnormal diffusion rate calculation: the measured or recorded real-time wind speed is input into the wind factor interference model to obtain the theoretical abnormal diffusion rate; actual and theoretical comparison: the theoretical diffusion rate is compared with the actual abnormal diffusion rate of the interference point to evaluate the accuracy of the model; model update decision: whether to update the wind factor interference model based on the comparison results, where the update includes adjusting the model parameters or retraining the model; specifically, assuming that When the diffusion direction is from water area to land area, multiple detection points include eight azimuth points (B1, B2, ..., B8) located around the water area. At this time, the wind direction is facing B1, so B1 is defined as the interference point, and the remaining 7 detection points are analysis points. The wind speed is obtained and substituted into the wind factor interference model to obtain a theoretical diffusion rate. The diffusion rate of the actual point B1 is then calculated and compared. If it is within the predetermined range, it can be ignored and determined to be without deviation. At this time, the wind factor interference model is determined to be accurate. The remaining 7 analysis points can be directly substituted into the model according to the wind speed intensity at the detection point to obtain data. There is no need to calculate and analyze all detection points one by one, which reduces the amount of calculation and improves the monitoring efficiency of the diffusion rate.

[0038] In addition, the number and location of detection points can be adjusted according to actual needs. If the diffusion direction is from the region to the water area, only one detection point can be set because the pollution disperses and dilutes quickly in the water. However, in order to ensure accurate detection, multiple detection points can be set for verification and monitoring.

[0039] like Figure 2As shown in the figure, the dynamic data acquisition module currently uses drones for environmental inspection or data collection, which is a common method. Therefore, the video images of the area collected by drones equipped with high-definition cameras are processed by conventional image processing and semantic segmentation based on the video graphics to obtain an accurate image of the area, which is used as the vegetation cover image.

[0040] It also includes an abnormal factor feedback module. Before this, the area is divided according to each detection point. For example, if there are 8 detection points, the area is divided into 8 areas. The vegetation coverage rate in the area of ​​each detection point is analyzed by color according to the vegetation coverage image. The vegetation coverage rate of each detection point is compared with the coverage threshold. According to the comparison result, it is judged whether there is vegetation loss anomaly at the corresponding detection point. If so, the detection point with vegetation loss anomaly is defined as a loss point, and the remaining detection points are defined as normal points. The marking factor of the loss point is then compared with the marking factor of the normal point. If the marking factor of the loss point is greater than the marking factor of the normal point, an instruction corresponding to vegetation loss and abnormal factor is output. Specifically, the marking factors of the loss point and the normal point are compared (which may refer to various environmental factors affecting vegetation growth, such as soil moisture, light intensity, pollution source composition, etc.). If the marking factor of the loss point is significantly higher than that of the normal point, this may indicate that these factors are directly related to vegetation loss. Its main purpose is to be able to determine whether the impact of the abnormal factor on vegetation loss corresponds. If so, the manager or automation system is prompted to take targeted solutions. If not, the instruction of vegetation loss to be determined is output, and further investigation is required.

[0041] The anomaly factor feedback module first performs semantic segmentation: based on the vegetation cover image, the pre-trained semantic segmentation model is used to segment the vegetation covered area and the vegetation missing area in the region; contour detection: the contour detection algorithm (Canny edge detection, Hough transform, etc.) is then used to perform contour planning on the vegetation covered area and the vegetation missing area to obtain a regional contour map; mapping and analysis: each detection point is mapped to the regional contour map, and the number and type of marker factors falling into the vegetation missing area are recorded, which helps to identify which anomaly factors are more significant in a specific area. Interference factor screening: If multiple (such as greater than or equal to 2) different types of marker factors are detected in a certain area, the system may believe that there is mutual interference between these factors, and output an interference factor screening instruction, indicating that more complex analysis or experiments are needed to determine the main influencing factors. Conversely, if only one or a few marker factors are detected in the area, and these factors are highly correlated with vegetation missing, the system can confirm that these factors are anomaly factors of vegetation missing and output corresponding processing suggestions. The type of output marker factor is anomaly factor of vegetation missing.

[0042] The detection part includes an insertion rod, which includes at least two rod sections. The two adjacent rod sections can rotate and extend. Specifically, the principle of rotation and extension is to control the rotation of the rod with the gas sensor through a motor, gear transmission or other mechanical structure. The two rod sections are threadedly connected to each other to achieve rotation and extension. The outer surface of each rod section is provided with an opening that passes through the rod section, and a through channel 5 is formed in the opening. A number of detection sensors are provided in the through channel 5.

[0043] like Figure 3 As shown, the rod is a water area rod, which is mainly used in water areas. The water area rod includes three rod sections, which are the first rod section 1, the second rod section 2 and the third rod section 3 from bottom to top. The first rod section 1 is provided with a mud discharge hole 4 at one end away from the second rod section 2. The mud discharge hole 4 is located on the outer surface of the first rod section 1, and the mud discharge hole 4 is communicated with the through-channel 5. In the initial state, the through-channel 5 of the first rod section 1 is located in the second rod section 2. In the extended state, the through-channel 5 of the first rod section 1 is exposed. An integrated soil detection sensor is provided in the through-channel 5 of the first rod section 1, an integrated water quality detection sensor is provided in the through-channel 5 of the second rod section 2, and an integrated gas detection sensor is provided in the through-channel 5 of the third rod section 3. Specifically, when the water area rod is first inserted into the water area, mud is filled in from the through-channel 5 of the first rod section 1, and the soil The detection sensor can detect the composition and content of the silt filled in the through channel 5. The normal water quality sensor can detect factors such as the composition in the water. The third section of the rod 3 is exposed above the water surface, and the through channel 5 on the third section is located above the water surface. It also includes a second adjustment module. When it is necessary to obtain data on sediment in the water area at the next time point, the first section of the rod 1 is controlled to extend outward from the second section of the rod 2, so that the part of the through channel 5 located in the second section of the rod 2 is extended, that is, the first section of the rod 1 will be telescoped downward into the silt. At this time, the through channel 5 of the first section of the rod 1 extends downward. At this time, there will be some space in the first rod, and the silt enters from the through channel 5 and then overflows from the mud discharge hole 4, or the mud discharge hole 4 is located above the through channel 5, and the silt enters from the mud discharge hole 4 and then is discharged from the through channel 5. At this time, the silt detected by the soil detector is new silt.

[0044] In addition, when the first section of the rod 1 is inserted into the silt, the silt enters directly from the through channel 5. At this time, the first section of the rod 1 extends downward, the amount of silt entering the through channel 5 increases, and some old mud overflows from the mud discharge hole 4 below. At this time, the silt detected by the soil detector is new silt.

[0045] The insertion rod is a field insertion rod, which is mainly used in the field. The field insertion rod includes two sections of rods. The end of any section of the rod away from the other section of the rod is a sharp end. The field insertion rod also includes a first section and a second section. The first section is the ground insertion section, which is mainly used to detect the components in the soil, and the second section is used to detect the components in the air.

[0046] Based on the above-mentioned water area insertion rod or regional insertion rod, the principle of air detection is the same, and it also includes a first adjustment module to obtain the real-time wind direction measured or recorded externally as the detection wind direction, and control the rotation of the rod with the gas sensor so that the through-channel 5 of the rod is aligned with the wind direction. The purpose is that different wind direction angles can easily blow the air over the water area to the region, and when the wind direction is completely opposite to the direction of the through-channel 5, the air cannot smoothly enter the through-channel 5, and the wind direction is an important factor affecting the distribution and diffusion of gases in the air. By aligning the through-channel 5 of the gas sensor with the wind direction, it can be ensured that the sensor is directly exposed to the gas flow brought by the dominant wind, thereby more accurately capturing the concentration and composition of the target gas. This helps to reduce detection errors caused by wind direction deviation and improve the accuracy and reliability of the data.

[0047] Furthermore, under certain wind conditions, certain gas components in the air may travel at higher concentrations or faster speeds. By adjusting the orientation of the gas sensor to align with the wind direction, these gas components can be captured more effectively, shortening detection time and improving detection efficiency.

[0048] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that improvements and modifications that do not depart from the principles of the present invention are within the scope of protection of the present invention.

Claims

1. An ecological environment monitoring and feedback system, characterized by: include Dynamic data acquisition module, which obtains real-time data collected from water areas and test pieces in the region as dynamic analysis factors; The module for comparing similar data compares the dynamic analysis factors corresponding to the water area with those corresponding to the region, or compares them with similar standard thresholds, and generates a dynamic factor comparison table based on the comparison results; The marker data analysis module selects an abnormal factor from the dynamic factor comparison table as a marker factor, calculates the marker factor content in the water area and the marker factor content in the region at two different time points at a certain time, and then obtains the abnormal diffusion rate of the marker factor in the water area and the abnormal diffusion rate of the marker factor in the region at the two time points through diffusion calculation, and then determines the diffusion direction based on the abnormal diffusion rate, wherein the diffusion direction includes diffusion from the water area to the region or diffusion from the region to the water area; An early warning decision output module indexes the exception handling solution corresponding to the exception type in the decision database according to the anomaly diffusion rate and diffusion direction; The detection member comprises an insertion rod, the insertion rod comprises at least two rod sections, two adjacent rod sections are capable of rotating and extending, an opening penetrating the rod section is provided on the outer surface of each rod section, a through channel (5) is formed in the opening, and a plurality of detection sensors are provided in the through channel (5); When the rod is a water area rod, the water area rod comprises three rod sections, namely a first rod section (1), a second rod section (2) and a third rod section (3); the first rod section (1) is provided with a mud discharge hole (4) at one end away from the second rod section (2); the mud discharge hole (4) is located on the outer surface of the first rod section (1); the mud discharge hole (4) is communicated with a through-channel (5); a soil quality detection sensor is provided in the through-channel (5) of the first rod section (1); a water quality detection sensor is provided in the through-channel (5) of the second rod section (2); and a water quality detection sensor is provided in the through-channel (5) of the third rod section (3). A gas detection sensor is provided in the channel (5); when the rod is a regional rod, the regional rod comprises two rod sections, and the end of any rod section away from the other rod section is a sharp end; it also includes a first adjustment module, which obtains the real-time wind direction measured or recorded externally as the detection wind direction, and controls the rotation of the rod with the gas sensor so that the through channel (5) of the rod is aligned with the wind direction; it also includes a second adjustment module, which controls the first rod section (1) to extend outward from the second rod section (2) when it is necessary to obtain data on sediment in the water area at the next time point, so that the silt at the bottom of the water area enters through the through channel (5).

2. The ecological environment monitoring and feedback system according to claim 1, characterized in that: It also includes a model building module and a model updating module. The model building module obtains historical data and performs training, testing and verification in a certain proportion to obtain a wind factor interference model; The model update module selects detection points with the same direction as the diffusion direction based on the measured or recorded real-time wind direction and defines them as interference points, defines the remaining detection points as analysis points, inputs the measured or recorded real-time wind speed into the wind factor interference model to obtain a theoretical abnormal diffusion rate, compares the theoretical diffusion rate with the actual abnormal diffusion rate of the interference point, and selects whether to update the wind factor interference model based on the comparison result.

3. The ecological environment monitoring and feedback system according to claim 2, characterized in that: When the diffusion direction is from water area to land area, the multiple detection points include eight azimuth points located around the water area.

4. The ecological environment monitoring and feedback system according to claim 2, characterized in that: The dynamic data acquisition module acquires video images of the area collected by the drone as vegetation cover images; It also includes an abnormal factor feedback module, which analyzes the vegetation coverage rate of each detection point through color analysis based on the vegetation coverage image, compares the vegetation coverage rate of each detection point with the coverage threshold, and judges whether there is a vegetation missing abnormality at the corresponding detection point based on the comparison result. If so, the detection point with the vegetation missing abnormality is defined as a missing point, and the remaining detection points are defined as normal points. The marking factor of the missing point is compared with the marking factor of the normal point. If the marking factor of the missing point is greater than the marking factor of the normal point, the instruction corresponding to the vegetation missing and abnormal factor is output; otherwise, the instruction for vegetation missing to be determined is output.

5. The ecological environment monitoring and feedback system according to claim 4, characterized in that: The abnormal factor feedback module divides the vegetation coverage area and the vegetation missing area in the region through the semantic segmentation model according to the vegetation coverage image, and then uses the contour detection algorithm to perform contour planning on the vegetation coverage area and the vegetation missing area to obtain a regional contour map, maps each detection point into the regional contour map, and records the number and type of marking factors falling into the vegetation missing area. According to the number of the marking factors, if it is greater than or equal to 2, the interference factor screening instruction is output; otherwise, the type of the output marking factor is an abnormal factor of vegetation missing.

Images