Vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis

Abstract

The invention relates to a vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis, and belongs to the technical field of ecological environment monitoring. The system comprises a satellite remote sensing data acquisition module, a data storage module used for storing satellite remote sensing data, an AI intelligent analysis module used for detecting vegetation coverage change conditions and disaster occurrence conditions according to the satellite remote sensing data, and a warning module used for giving a warning according to the disaster occurrence conditions. The park scenic region ecological change monitoring system is built based on satellite remote sensing and AI intelligent analysis, comprehensive, accurate and real-time monitoring of the ecological condition of the park scenic region is achieved, region division is conducted on the park scenic region through the disaster analysis module, the path planning module gives out an evacuation path according to the position relation between a road section and a disaster region and an isolation belt, and the evacuation path planning module gives out evacuation information according to the position relation between the road section and the isolation belt. Therefore, the evacuation module is guided to lead victims to go to a safe area.

Description

Technical Field

[0001] The present invention belongs to the technical field of ecological environment monitoring, and specifically relates to a vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis. Background Art

[0002] National parks are important areas for protecting natural ecosystems and maintaining biodiversity. With the increase of human activities and the impact of climate change, the ecological environment in national parks faces many challenges, such as habitat loss, invasion of alien species, and the impact of climate change on species distribution and ecological processes. In order to effectively protect these key ecosystems, it is urgent to monitor the ecological status of national parks continuously, comprehensively and accurately. Among them, vegetation coverage reflects the health status, biodiversity and land use of national park ecosystems.

[0003] Traditional ecological monitoring methods mainly rely on ground observations and field surveys. Although these methods can provide a certain degree of accuracy, they have limited coverage, high costs, and low time efficiency when facing large national parks. With the development of technology, satellite remote sensing technology has become an important tool for ecological monitoring due to its large-scale, all-weather, and multi-temporal characteristics. However, the complexity of satellite data processing and analysis, as well as the huge amount of data, have brought great challenges to traditional analysis methods. Artificial intelligence technology, especially machine learning and deep learning algorithms, provides a new way to process and analyze big data, which can significantly improve monitoring efficiency and analysis accuracy. However, the existing national park monitoring system is difficult to make full use of the above technologies to effectively monitor and protect the changes in vegetation coverage and disasters in national parks. Summary of the invention

[0004] The purpose of the present invention is to provide a vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis in order to solve the above problems, aiming to provide an efficient, accurate and automated ecological monitoring solution to comprehensively monitor the ecological health status of national parks.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] A vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis, which is used to monitor the vegetation coverage of park scenic spots, including:

[0007] Satellite remote sensing data acquisition module, used to obtain satellite remote sensing data of park scenic spots;

[0008] A data storage module, used for storing the satellite remote sensing data;

[0009] An AI intelligent analysis module is used to construct a network model based on the satellite remote sensing data and detect changes in vegetation coverage and disaster occurrence; wherein the network model includes a vegetation image recognition network model and a disaster image recognition network model;

[0010] An alarm module, used for issuing an alarm according to the occurrence of the disaster;

[0011] Among them, the AI ​​intelligent analysis module includes an image preprocessing module, a coverage analysis module and a disaster analysis module. During the monitoring process, the coverage analysis module detects the vegetation coverage change rate in the park scenic area based on satellite remote sensing data and the vegetation image recognition network model. When the vegetation coverage change rate exceeds the set value, the disaster analysis module detects whether a disaster occurs in the park scenic area based on the satellite remote sensing data and the disaster image recognition network model. When a disaster occurs, the warning module warns the affected people.

[0012] As a further optimization scheme of the present invention, the warning module includes a flying guidance and evacuation module, a digital display module and an alarm module; when the disaster analysis module detects that no disaster has occurred in the park scenic area, the satellite remote sensing data and the vegetation coverage change rate are presented through the digital display module; when the disaster analysis module detects that a disaster has occurred in the park scenic area, an alarm is sounded through the alarm module, and the affected people in the park scenic area are evacuated to a safe area through the guidance and evacuation module.

[0013] As a further optimization scheme of the present invention, the AI ​​intelligent analysis module also includes an image post-processing module and a path planning module. The disaster analysis module divides the park scenic area into regions according to satellite remote sensing data to obtain disaster boundaries. The image post-processing module analyzes the warning boundaries according to the disaster boundaries. The area enclosed by the disaster boundaries is the disaster area, and the area between the disaster boundary and the warning boundary is the isolation zone. The path planning module sets a patrol track and an evacuation path according to the disaster area and the isolation zone. The evacuation guidance module can search for disaster victims in the park scenic area along the patrol track, and can evacuate the disaster victims along the evacuation path.

[0014] As a further optimization solution of the present invention, the evacuation guidance module divides the road section into dangerous sections and safe sections according to the disaster area and the isolation zone, and the evacuation route is composed of several safe sections.

[0015] As a further optimization scheme of the present invention, there are multiple guiding evacuation modules. After the disaster analysis module detects the occurrence of a disaster, some of the guiding evacuation modules are in patrol mode and fly along each patrol trajectory, and the other part of the guiding evacuation modules are in standby mode. When the guiding evacuation module in patrol mode searches for disaster victims, it enters evacuation mode and guides the disaster victims to a safe area. Whenever the working mode of one guiding evacuation module is changed from patrol mode to evacuation mode, the working mode of another guiding evacuation module is changed from standby mode to patrol mode.

[0016] As a further optimization scheme of the present invention, when the working mode of the evacuation guidance module is the evacuation mode, the AI ​​intelligent analysis module updates the disaster area and isolation zone in real time. When the evacuation path of the evacuation guidance module passes through the updated disaster area or isolation zone, the path planning module re-plans the evacuation path.

[0017] As a further optimization scheme of the present invention, the evacuation guidance module includes a drone and a camera module, a human image recognition module and a voice broadcast module arranged on the drone. The camera module is used to collect post-disaster image data, the human image recognition module is used to identify disaster victims in combination with the post-disaster image data, and the voice broadcast module is used to issue follow-up prompts to the disaster victims.

[0018] As a further optimization scheme of the present invention, the evacuation guidance module also includes a position analysis module, which determines whether the drone is far away from the disaster victims based on the recognition results of the human image recognition module. When the drone is far away from the disaster victims, the drone is put into a hovering state.

[0019] The beneficial effects of the present invention are:

[0020] 1) The present invention has built a park scenic area ecological change monitoring system based on satellite remote sensing and AI intelligent analysis. By integrating satellite remote sensing technology and artificial intelligence algorithms, it has achieved comprehensive, accurate and real-time monitoring of the ecological status of park scenic areas. The system can provide managers with scientific decision-making support, effectively respond to ecological threats, and promote ecological protection and sustainable management of national parks;

[0021] 2) The AI ​​intelligent analysis module of the present invention includes a coverage analysis module and a disaster analysis module. The coverage analysis module detects the vegetation coverage change rate in the park scenic area based on satellite remote sensing data and a vegetation image recognition network model. When the change rate exceeds a set value, the disaster analysis module detects whether a disaster has occurred in the park scenic area based on satellite remote sensing data and a disaster image recognition network model. It can monitor the vegetation coverage in real time, and activate the disaster analysis module based on the real-time monitoring results to reduce invalid monitoring of the disaster analysis module;

[0022] 3) The present invention divides the park scenic area into regions through the disaster analysis module to obtain the disaster boundary, and the disaster boundary is offset outward through the image post-processing module to obtain the warning boundary, and the disaster area and the isolation zone are divided based on this, and a number of patrol tracks are set layer by layer around the geometric center of the disaster area. The guidance evacuation module can search for disaster victims in the park scenic area along the patrol tracks. When the guidance evacuation module searches for disaster victims, the path planning module gives an evacuation path according to the positional relationship between each road section and the disaster area and the isolation zone, so that the guidance evacuation module can lead the disaster victims to a safe area;

[0023] 4) After the disaster analysis module of the present invention detects the occurrence of a disaster, a part of the evacuation guidance modules are in patrol mode and fly along each patrol track respectively, and another part of the evacuation guidance modules are in standby mode and are located inside the innermost patrol track. When the evacuation guidance modules in patrol mode search for disaster victims, they enter evacuation mode and guide the disaster victims to a safe area. Whenever the working mode of one evacuation guidance module is changed from patrol mode to evacuation mode, the working mode of another evacuation guidance module is changed from standby mode to patrol mode, so as to ensure the orderliness of patrol search work and evacuation guidance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0025] Figure 2 is a system block diagram of the present invention;

[0026] Figure 3 This is a schematic diagram of the structure of the guiding evacuation module of the present invention.

[0027] Figure 4 It is a schematic diagram of a park scenic area of ​​the present invention.

[0028] Figure 5 It is a schematic diagram of the working status of the guiding evacuation module in the patrol mode of the present invention.

[0029] Figure 6 for Figure 5 Enlarged view of point A in the middle.

[0030] In the figure: S, road section; P, bifurcation node; B1, disaster boundary; B2, warning boundary; D, disaster area; Q, isolation zone; C, geometric center; T, patrol trajectory. DETAILED DESCRIPTION

[0031] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0032] Example

[0033] like Figure 1 and Figure 2 As shown, this embodiment relates to a vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis, including a satellite remote sensing data acquisition module for acquiring satellite remote sensing data of a park scenic area, a data storage module for storing satellite remote sensing data, an AI intelligent analysis module for building a network model based on satellite remote sensing data, and an alarm module for issuing an alarm based on the occurrence of a disaster. The AI ​​intelligent analysis module is arranged in a monitoring center, and the AI ​​intelligent analysis module is also used to detect the vegetation coverage change and disaster occurrence in the park scenic area. The network model includes a vegetation image recognition network model and a disaster image recognition network model. The vegetation image recognition network model is a generalized regression neural network model. The disaster image recognition network model is a MobileCBAM-Net network model, i.e., a mobile convolutional block attention network model. Among them, the AI ​​intelligent analysis module includes an image preprocessing module, a coverage analysis module, and a disaster analysis module.

[0034] The monitoring system of this embodiment is mainly applicable to park scenic spots such as national parks. The ecological environment in the park scenic spots faces many challenges, such as habitat loss, invasion of alien species, and the impact of climate change on species distribution and ecological processes. In order to effectively protect these key ecosystems, it is urgent to monitor the ecological conditions of national parks continuously, comprehensively and accurately. The disaster in this embodiment is fire. In some other embodiments, a network model that can monitor disasters such as mudslides and floods can also be constructed by replacing the training set and the test set.

[0035] During the monitoring process, the satellite remote sensing data is preprocessed by the image preprocessing module to obtain the image data to be measured. The above preprocessing includes radiation correction, geometric correction, noise filtering and other operations. The preprocessing process is a conventional means of processing remote sensing data in this field, so it will not be described in detail. After preprocessing, the coverage analysis module is based on the satellite remote sensing data and the vegetation image recognition network model. Specifically, the coverage analysis module inputs the image data to be measured into the vegetation image recognition network model, analyzes the vegetation coverage through the vegetation image recognition network model, and the data storage module can store the historical data of vegetation coverage. The coverage analysis module calculates the change rate of vegetation coverage in the park scenic area based on the current vegetation coverage and the historical data in the data storage module. The specific vegetation coverage analysis process can refer to the patent document with publication number CN104142142B. When the vegetation coverage change rate exceeds the set value, the disaster analysis module detects whether a disaster occurs in the park scenic area based on the satellite remote sensing data and the disaster image recognition network model. Specifically, the disaster analysis module inputs the image data to be tested into the disaster image recognition network model, and identifies the fire through the disaster image recognition network model. The specific process of fire image recognition is common knowledge in this field, so it will not be described in detail. When a disaster occurs in the park scenic area, the warning module warns the victims.

[0036] Specifically, the warning module includes a flying guidance evacuation module, a digital display module and an alarm module, wherein the guidance evacuation module is arranged in the drone site, and the digital display module and the alarm module are arranged in the monitoring center. When the disaster analysis module detects that no disaster has occurred in the park scenic area, the satellite remote sensing data and the vegetation coverage change rate are presented through the digital display module. When the disaster analysis module detects that a disaster has occurred in the park scenic area, in addition to presenting the satellite remote sensing data and the vegetation coverage change rate through the digital display module, an alarm is also issued through the alarm module, and the disaster victims in the park scenic area are evacuated to a safe area through the guidance evacuation module. The alarm module can issue an alarm to notify the relevant management department by SMS, email, etc. In addition, when the vegetation coverage change rate exceeds the set value, the phenomenon may also be caused by illegal logging. In some other embodiments, cameras can be evenly arranged around the park scenic area. When the disaster analysis module analyzes that there is a natural disaster in the park scenic area, the cameras around the park scenic area are turned on, and the cameras transmit the image to the digital display module of the monitoring center in real time for the staff to view and find the person who illegally cuts down trees.

[0037] See also Figure 4-6, where the dense dotted line is used to represent the road section S, and the sparse dotted line is used to represent the patrol track T. In order to ensure the clarity of the attached figure, only some sections in the park scenic area are shown in the figure. The AI ​​intelligent analysis module also includes an image post-processing module and a path planning module. While detecting the occurrence of disasters, the disaster analysis module divides the park scenic area into regions according to the image data to be measured formed by satellite remote sensing data to obtain the disaster boundary B1. The image post-processing module offsets the disaster boundary B1 outward to obtain the warning boundary B2. The area surrounded by the disaster boundary B1 is the disaster area D, and the area between the disaster boundary B1 and the warning boundary B2 is the isolation belt Q. The width of the isolation belt Q can be set according to the actual situation. The specific process of the above-mentioned area division can refer to the patent document with the publication number CN114821289B. In the process of offsetting the disaster boundary B1, when there is a part in the disaster boundary B1 that has too large a curvature and cannot be offset smoothly, the image post-processing module uses a smoothing technology such as Gaussian filtering to process the image, and then offsets the disaster boundary B1. The path planning module is used to analyze the geometric center C of the disaster area D, and to set a number of patrol tracks T layer by layer around the geometric center C. The evacuation guidance module can search for disaster victims in the park scenic area along the patrol track T. When the evacuation guidance module finds the disaster victims, the path planning module gives an evacuation path. The geometric center C is preferably the centroid of the disaster area D. In this embodiment, the patrol track T is circular. In some other embodiments, the shape of the patrol track T can also be an ellipse, a crescent, a star, etc., and the patrol track T of the corresponding shape can also be set according to the shape of the disaster area D to improve the search efficiency.

[0038] There are multiple road sections S and several bifurcated nodes P in the park scenic area. The road sections S are formed between adjacent bifurcated nodes P and between the bifurcated nodes P close to the edge of the park scenic area and the edge of the park scenic area. The road sections S are divided into dangerous sections and safe sections. Among them, the dangerous sections are the sections in the park scenic area that pass through the isolation belt Q or the disaster area D, and the safe sections are the remaining sections in the park scenic area. The evacuation path planned by the path planning module is formed by splicing several safe sections. It should be noted that when the victims are in the dangerous section, the section where the starting point of the evacuation path is located is the safe section closest to the victims. When a fire occurs, the area in the park scenic area that lacks flammable materials or has sound fire-fighting facilities is a safe area. In the process of planning the evacuation route, it is first determined whether the disaster victims are in a safe section. If the disaster victims are in a safe section, the path planning module calculates the shortest path from the disaster victims to each safe area in turn, and then selects a shortest path from the shortest paths corresponding to different safe areas as the evacuation path; if the disaster victims are in a dangerous section, the path planning module first analyzes the safe section closest to the disaster victims, takes the end of the safe section close to the disaster victims as the starting point, calculates the shortest path from the starting point to each safe area in turn, and then selects a shortest path from the shortest paths corresponding to different safe areas as the evacuation path, guiding the evacuation module to first lead the disaster victims to the starting point, and then evacuate the disaster victims along the evacuation path.

[0039] The AI ​​intelligent analysis module also includes a state control module. There are multiple guidance and evacuation modules. After the disaster analysis module detects that a disaster has occurred, the state control module controls each guidance and evacuation module to fly away from the drone site. After that, some of the guidance and evacuation modules are in patrol mode and fly along each patrol track T respectively. The other part of the guidance and evacuation modules are in standby mode and are located inside the innermost patrol track T. When the guidance and evacuation modules in patrol mode search for disaster victims, they send feedback signals to the state control module and enter the evacuation mode to guide the disaster victims to a safe area. Each guidance and evacuation module in patrol mode is evenly distributed on each patrol track T. The number of guidance and evacuation modules on the patrol track T is positively correlated with the path length of the patrol track T. Whenever the working mode of a guidance and evacuation module changes from patrol mode to evacuation mode, the state control module controls the working mode of another guidance and evacuation module to change from standby mode to patrol mode, and replaces the guidance and evacuation module whose working mode changes to evacuation mode to perform patrol search work. After the guidance and evacuation module leads the disaster victims to a safe area, it enters the innermost patrol track T and enters standby mode. When the evacuation guidance module is in evacuation mode, the AI ​​intelligent analysis module updates the disaster area D and isolation zone Q in real time. When the evacuation path of the evacuation guidance module passes through the updated disaster area D or isolation zone Q, the path planning module replans the evacuation path. In addition, there are several landmarks in the park scenic area. The AI ​​intelligent analysis module also includes a positioning module. The positioning module determines the geographical location of the disaster area D and the isolation zone Q based on the pixels corresponding to the landmarks and the pixels corresponding to the disaster area D, so that the path planning module can plan the evacuation path.

[0040] The guidance evacuation module includes a drone and a camera module, a human image recognition module and a voice broadcast module arranged on the drone. The camera module is used to collect post-disaster image data, the human image recognition module is used to identify the disaster victims in combination with the post-disaster image data, and the voice broadcast module is used to issue a follow-up prompt to the disaster victims. The guidance evacuation module also includes a position analysis module. The position analysis module determines whether the drone is far away from the disaster victims based on the recognition result of the human image recognition module. When the drone is far away from the disaster victims, the drone is put into a hovering state. Human image recognition is common knowledge in the art, so it is not described in detail here. The position analysis module specifically determines whether the drone is far away from the disaster victims based on the position of the corresponding pixels of the disaster victims relative to the post-disaster image collected by the camera module. In the specific judgment process, when the corresponding pixels of the disaster victims are close to the edge of the post-disaster image, it is judged that the drone is far away from the disaster victims. In addition, the monitoring system of this embodiment also includes a disaster reduction module. The disaster reduction module is a drone equipped with fire extinguishing equipment and a camera. While evacuating the disaster victims, the disaster area D can be extinguished to prevent the fire from spreading rapidly.

[0041] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis, which is used to monitor the vegetation coverage of park scenic spots, and is characterized by: include: Satellite remote sensing data acquisition module, used to obtain satellite remote sensing data of park scenic spots; A data storage module, used for storing the satellite remote sensing data; An AI intelligent analysis module is used to construct a network model based on the satellite remote sensing data and detect changes in vegetation coverage and disaster occurrence; wherein the network model includes a vegetation image recognition network model and a disaster image recognition network model; An alarm module, used for issuing an alarm according to the occurrence of the disaster; Among them, the AI ​​intelligent analysis module includes an image preprocessing module, a coverage analysis module and a disaster analysis module. During the monitoring process, the coverage analysis module detects the vegetation coverage change rate in the park scenic area based on satellite remote sensing data and the vegetation image recognition network model. When the vegetation coverage change rate exceeds the set value, the disaster analysis module detects whether a disaster occurs in the park scenic area based on the satellite remote sensing data and the disaster image recognition network model. When a disaster occurs, the warning module warns the affected people.

2. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 1 is characterized by: The warning module includes a flying guidance and evacuation module, a digital display module and an alarm module; when the disaster analysis module detects that no disaster has occurred in the park scenic area, the satellite remote sensing data and the vegetation coverage change rate are presented through the digital display module; when the disaster analysis module detects that a disaster has occurred in the park scenic area, an alarm is sounded through the alarm module, and the affected people in the park scenic area are evacuated to a safe area through the guidance and evacuation module.

3. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 2 is characterized by: The AI ​​intelligent analysis module also includes an image post-processing module and a path planning module. The disaster analysis module divides the park scenic area into regions according to satellite remote sensing data to obtain disaster boundaries. The image post-processing module analyzes the warning boundaries according to the disaster boundaries. The area enclosed by the disaster boundaries is the disaster area, and the area between the disaster boundary and the warning boundary is the isolation zone. The path planning module sets a patrol track and an evacuation path according to the disaster area and the isolation zone. The evacuation guidance module can search for disaster victims in the park scenic area along the patrol track and can evacuate the disaster victims along the evacuation path.

4. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 3 is characterized by: The evacuation guidance module divides the road section into a dangerous section and a safe section according to the disaster area and the isolation zone, and the evacuation route is composed of a number of safe sections.

5. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 4 is characterized in that: There are multiple guiding evacuation modules. After the disaster analysis module detects the occurrence of a disaster, some of the guiding evacuation modules are in patrol mode and fly along patrol tracks respectively, while other guiding evacuation modules are in standby mode. When the guiding evacuation modules in patrol mode search for disaster victims, they enter evacuation mode and guide the disaster victims to a safe area. Whenever the working mode of one guiding evacuation module is changed from patrol mode to evacuation mode, the working mode of another guiding evacuation module is changed from standby mode to patrol mode.

6. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 5 is characterized by: When the working mode of the evacuation guidance module is the evacuation mode, the AI ​​intelligent analysis module updates the disaster area and isolation zone in real time. When the evacuation path of the evacuation guidance module passes through the updated disaster area or isolation zone, the path planning module re-plans the evacuation path.

7. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 5 is characterized by: The evacuation guidance module includes a drone and a camera module, a human image recognition module and a voice broadcast module arranged on the drone. The camera module is used to collect post-disaster image data, the human image recognition module is used to identify disaster victims in combination with the post-disaster image data, and the voice broadcast module is used to issue follow-up prompts to the disaster victims.

8. The vegetation coverage change monitoring system based on satellite remote sensing and AI intelligent analysis according to claim 7 is characterized by: The evacuation guidance module also includes a position analysis module, which determines whether the drone is far away from the disaster victims based on the recognition result of the human image recognition module, and when the drone is far away from the disaster victims, the drone is put into a hovering state.

Images