A method and device for monitoring wetland vegetation degradation based on satellite remote sensing images

Abstract

The present invention provides a wetland vegetation degradation monitoring method and device based on satellite remote sensing images, wherein the method comprises: obtaining remote sensing images of a monitoring area, and calculating reflectivity data of the monitoring area according to the remote sensing images; performing layered processing on the remote sensing images based on the reflectivity data, and filtering interference items to obtain layered images, wherein the interference items include cloud layers, ice layers, and snow layers; obtaining the spatial bottom layer of the monitoring area according to the layered images, wherein the spatial bottom layer includes a wetland vegetation layer; performing point cloud coordinate layout on the wetland vegetation layer through a preset point cloud model to generate a data floating point layer; performing cloud coordinate comparison on the data floating point layer and the historical data floating point layer of the previous cycle according to the monitoring cycle to obtain a differentiated value, and evaluating the differentiated value to obtain a wetland vegetation degradation evaluation result. The present invention can realize high-precision automatic monitoring of wetland vegetation, improve monitoring efficiency, and help to timely discover the trend of wetland vegetation degradation.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing data monitoring, and in particular to a wetland vegetation degradation monitoring method, device, electronic equipment and computer-readable storage medium based on satellite remote sensing images. Background Art

[0002] With the continuous development and utilization of the natural environment, wetland vegetation, as an important component of the ecosystem, plays a vital role in maintaining ecological balance and biodiversity. However, in recent years, due to climate change, human interference and other factors, wetland vegetation degradation has become increasingly serious, and an effective monitoring method is urgently needed to timely detect and evaluate the degradation of wetland vegetation.

[0003] Traditional wetland vegetation monitoring methods mainly rely on manual surveys and ground observations, which are not only time-consuming and labor-intensive, but also difficult to cover large areas and cannot meet the needs of real-time monitoring and comprehensive assessment. With the development of satellite remote sensing technology, monitoring wetland vegetation degradation through satellite remote sensing images has become a feasible method.

[0004] At present, although there are some wetland vegetation monitoring methods based on satellite remote sensing, they are often affected by interference factors such as clouds, ice and snow, which limits the accuracy and reliability of the monitoring results. In addition, the existing monitoring methods usually lack quantitative assessment of the tendency of wetland vegetation degradation, making it difficult to provide a scientific basis for wetland protection and management.

[0005] Therefore, there is an urgent need for a wetland vegetation degradation monitoring method that can realize high-precision automated monitoring of wetland vegetation, improve monitoring efficiency, and quantitatively evaluate the degradation tendency of wetland vegetation. Summary of the invention

[0006] Based on this, it is necessary to provide a wetland vegetation degradation monitoring method, device, electronic device and computer-readable storage medium based on satellite remote sensing images to address the above technical problems.

[0007] A wetland vegetation degradation monitoring method based on satellite remote sensing images comprises the following steps: acquiring a remote sensing image of a monitoring area, and calculating reflectivity data of the monitoring area based on the remote sensing image; performing layered processing on the remote sensing image based on the reflectivity data, and filtering interference items to obtain a layered image, wherein the interference items include cloud layers, ice layers, and snow layers; obtaining a spatial bottom layer of the monitoring area based on the layered image, wherein the spatial bottom layer includes a wetland vegetation layer; performing point cloud coordinate layout on the wetland vegetation layer through a preset point cloud model to generate a data floating point layer; performing cloud coordinate comparison on the data floating point layer and the historical data floating point layer of the previous period according to the monitoring period to obtain a differentiated value, and evaluating the differentiated value to obtain a wetland vegetation degradation assessment result.

[0008] In one of the embodiments, the remote sensing image of the monitoring area is obtained, and the reflectivity data of the monitoring area is calculated based on the remote sensing image, including: receiving the location information of the monitoring area, and synchronizing it to the satellite remote sensing system; obtaining the remote sensing image corresponding to the location information through the satellite remote sensing system, and obtaining the resolution of the monitoring area; obtaining each reflectivity in the remote sensing image at the same time, and calculating each reflectivity to obtain reflectivity data.

[0009] In one embodiment, the remote sensing image is layered based on the reflectivity data, and interference items are filtered to obtain a layered image, including: layering the remote sensing image based on the reflectivity data; obtaining preset bands, and filtering interference items of the layered remote sensing image according to the preset bands to obtain a layered image, wherein the preset bands include a green band, a blue band, a red band, a near-infrared band, a mid-infrared band, a first short-wave infrared band, and a second short-wave infrared band.

[0010] In one of the embodiments, obtaining the spatial bottom layer of the monitoring area based on the layered image includes: determining the spatial distance between the remote sensing image and the satellite based on the reflectivity data; and subtracting the spatial distance based on the reflectivity data to obtain the spatial bottom layer of the detection area.

[0011] In one of the embodiments, the training process of the point cloud model includes: obtaining a historical remote sensing image data set, performing image recognition on the images in the historical remote sensing image data set to obtain image elements, using a rendering model to perform model rendering training on the image elements to generate a three-dimensional rendering model, and the image elements include mountains, water, lakes, rivers and vegetation; using the three-dimensional rendering model to identify the color and shape of the images in the historical remote sensing image data set to obtain feature information; based on the feature information, marking the key features and contours of the image elements to generate a historical virtual line map; matching the historical virtual line map with the corresponding historical remote sensing image, evaluating the matching success rate, until the matching success rate reaches a preset threshold, and obtaining a trained point cloud model.

[0012] In one of the embodiments, the point cloud coordinates of the wetland vegetation layer are laid out through a preset point cloud model to generate a data floating-point layer, including: inputting the wetland vegetation layer into a preset point cloud model, and outputting a rendering modeling diagram, wherein the rendering modeling diagram includes a number of image elements; gridding the rendering modeling diagram to obtain a virtual line map, and forming point cloud coordinates corresponding to all image elements on the virtual line map based on the reflectivity data; creating floating points corresponding to the point cloud coordinates through the image frame number of the wetland vegetation layer, and combining with a reverse line map to generate a data floating-point layer, wherein the reverse line map is a line map with a reverse order to the virtual line map.

[0013] In one of the embodiments, according to the monitoring period, the data floating-point layer is compared with the historical data floating-point layer of the previous period in cloud coordinates to obtain differentiated values, and the differentiated values ​​are evaluated to obtain wetland vegetation degradation assessment results, including: according to the monitoring period, the historical data floating-point layer of the previous period is obtained, and placed in the same coordinate system as the current data floating-point layer; the coordinate positions of each floating point in the historical data floating-point layer and the data floating-point layer are identified, and the coordinate positions are classified based on the image elements to obtain classification results; the floating points in the historical data floating-point layer and the data floating-point layer are matched to obtain matching floating-point pairs; a connecting vector between two floating points is formed according to the matching floating-point pairs, and based on the classification results, the direction of the connecting vector is judged to be forward or reverse, and a differentiated value is obtained according to the connecting vector and its direction; based on the differentiated values ​​of all floating-point pairs, the degradation tendency of wetland vegetation is evaluated to obtain wetland vegetation degradation assessment results.

[0014] A wetland vegetation degradation monitoring device based on satellite remote sensing images is used to implement a wetland vegetation degradation monitoring method based on satellite remote sensing images as described above, comprising: a remote sensing image acquisition module, used to acquire remote sensing images of a monitoring area, and calculate reflectivity data of the monitoring area based on the remote sensing images; a remote sensing image stratification module, used to perform stratification processing on the remote sensing images based on the reflectivity data, and filter interference items to obtain stratified images, wherein the interference items include cloud layers, ice layers, and snow layers; a spatial bottom layer acquisition module, used to obtain the spatial bottom layer of the monitoring area based on the stratified images, wherein the spatial bottom layer includes a wetland vegetation layer; a point cloud coordinate layout module, used to perform point cloud coordinate layout on the wetland vegetation layer through a preset point cloud model to generate a data floating point layer; a vegetation degradation assessment module, used to perform cloud coordinate comparison on the data floating point layer with the historical data floating point layer of the previous period according to the monitoring period to obtain differential values, and evaluate the differential values ​​to obtain a wetland vegetation degradation assessment result.

[0015] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the wetland vegetation degradation monitoring method based on satellite remote sensing images described in the above-mentioned embodiments are implemented.

[0016] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the wetland vegetation degradation monitoring method based on satellite remote sensing images described in the above-mentioned embodiments.

[0017] Compared with the prior art, the advantages and beneficial effects of the present invention are: obtaining remote sensing images of the monitoring area, calculating the reflectivity data of the monitoring area, layering the remote sensing images based on the reflectivity data, filtering interference items such as cloud layers, ice layers and snow layers to obtain layered images, and ensuring the accuracy and reliability of the detection data by eliminating interference items, which is suitable for wetland vegetation monitoring in severe weather; obtaining the spatial bottom layer of the monitoring area according to the layered images, and the spatial bottom layer includes a wetland vegetation layer; performing point cloud coordinate layout of the wetland vegetation layer through a preset point cloud model to generate a data floating point layer, and according to the monitoring cycle, comparing the cloud coordinates of the data floating point layer with the historical data floating point layer of the previous cycle to obtain differentiated values, and evaluating the differentiated values ​​to obtain a wetland vegetation degradation assessment result, thereby realizing high-precision automated monitoring of wetland vegetation, improving monitoring efficiency, and being able to periodically evaluate the tendency of wetland vegetation degradation, which is helpful to timely discover the trend of wetland vegetation degradation and provide a scientific basis for wetland protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The figure is a flow chart of a method for monitoring wetland vegetation degradation based on satellite remote sensing images in one embodiment;

[0019] Figure 2 is a schematic structural diagram of a wetland vegetation degradation monitoring device based on satellite remote sensing images in one embodiment;

[0020] Figure 3 FIG. 1 is a schematic diagram of the internal structure of an electronic device in an embodiment. DETAILED DESCRIPTION

[0021] Before describing the specific embodiments of the present invention, the overall concept of the present invention is described as follows:

[0022] The present invention is mainly developed for the wetland vegetation degradation monitoring process. The existing monitoring methods are inefficient, easily interfered with, and difficult to achieve quantitative assessment of wetland vegetation degradation tendency.

[0023] Therefore, the present invention proposes a wetland vegetation degradation monitoring method based on satellite remote sensing images, which obtains remote sensing images of the monitoring area, calculates reflectivity data of the monitoring area, performs layered processing on the remote sensing images based on the reflectivity data, and filters interference items such as cloud layers, ice layers and snow layers to obtain layered images. By eliminating interference items, the accuracy and reliability of the detection data are ensured, and the method is suitable for wetland vegetation monitoring in severe weather; the spatial bottom layer of the monitoring area is obtained according to the layered images, and the spatial bottom layer includes a wetland vegetation layer; the wetland vegetation layer is laid out with point cloud coordinates through a preset point cloud model to generate a data floating-point layer, and according to the monitoring cycle, the data floating-point layer is compared with the historical data floating-point layer of the previous cycle for cloud coordinates to obtain differentiated values, and the differentiated values ​​are evaluated to obtain a wetland vegetation degradation assessment result, thereby realizing high-precision automated monitoring of wetland vegetation, improving monitoring efficiency, and being able to periodically evaluate the tendency of wetland vegetation degradation, which is helpful to timely discover the trend of wetland vegetation degradation and provide a scientific basis for wetland protection.

[0024] After introducing the overall concept of the present invention, in order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail by specific implementation methods in combination with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] In one embodiment, Figure 1 As shown, a wetland vegetation degradation monitoring method based on satellite remote sensing images is provided, comprising the following steps:

[0026] Step S101, obtaining a remote sensing image of a monitoring area, and calculating reflectivity data of the monitoring area according to the remote sensing image.

[0027] Specifically, the monitoring area where vegetation degradation monitoring is required is determined, the reflectance data of corresponding resolution within the monitoring area is obtained, and the remote sensing image of the monitoring area is obtained at the same time. The reflectance data is information describing the ratio of the surface reflectance of the object to the incident irradiance, which provides information on the surface cover and can be used to identify the objects.

[0028] Among them, step S101 includes: receiving the location information of the monitoring area and synchronizing it to the satellite remote sensing system; obtaining the remote sensing image corresponding to the location information through the satellite remote sensing system, and obtaining the resolution of the monitoring area; obtaining various reflectivities in the remote sensing image at the same time, and calculating various reflectivities to obtain reflectivity data.

[0029] Specifically, the user can obtain the location information of the monitoring area by preset or real-time input (software platform or web application). The location information may include latitude and longitude coordinates, place names or other data that can accurately locate the monitoring area, and synchronize the location information to the satellite remote sensing system to ensure that the satellite remote sensing system can accurately obtain the remote sensing image data of the monitoring area; after receiving the location information, the satellite remote sensing system adapts the most suitable resolution according to the characteristics and needs of the area. The size of the resolution will affect the clarity and information richness of the remote sensing image. According to the obtained resolution, the remote sensing image is adjusted accordingly to ensure the clarity and accuracy of the remote sensing image, which helps to improve the accuracy of subsequent analysis. The size of the remote sensing image is adjusted by resolution; at the same time, the reflectivity of various surface features in the remote sensing image is obtained. Reflectivity is the ability of surface objects to reflect light. Different types of surface covers, such as vegetation, water bodies or bare soil, have different reflectivity characteristics; the obtained reflectivities are calculated and processed to obtain accurate reflectivity data, which is convenient for subsequent surface cover classification, environmental monitoring and other analyses.

[0030] The calculation method of reflectivity data can adopt radiation calibration method, spatial resolution method, model inversion method and data fusion method, so as to facilitate the identification of ground objects based on reflectivity data.

[0031] Among them, the radiation calibration method is preferred. The radiation calibration method is a process of converting the digital quantization value of the remote sensing image into the surface reflectivity with actual physical meaning. It can eliminate the image of factors such as the sensor itself, atmospheric conditions, solar altitude angle and azimuth angle, so as to obtain more accurate surface reflectivity data.

[0032] Step S102, performing layered processing on the remote sensing image based on the reflectivity data and filtering interference items to obtain a layered image, wherein the interference items include cloud layers, ice layers, and snow layers.

[0033] Specifically, the remote sensing image is divided into different layers according to the reflectivity data obtained. At the same time, in order to improve the accuracy of the image, the divided layers also need to be filtered out of interference items such as clouds, ice and snow layers to obtain clear and accurate layered images.

[0034] Among them, step S102 includes: identifying different surface characteristics according to reflectivity data, and layering the remote sensing image according to the surface characteristics; obtaining preset bands, filtering interference items on the layered remote sensing image according to the preset bands, and obtaining a layered image, the preset bands include a green band, a blue band, a red band, a near-infrared band, a mid-infrared band, a first short-wave infrared band and a second short-wave infrared band.

[0035] Specifically, in remote sensing images, clouds, ice and snow often appear as interference items, affecting the accurate observation and analysis of the surface. For example, clouds have specific reflective properties in the visible and near-infrared bands. By identifying and processing the reflectivity data of these bands, the interference of clouds can be effectively removed.

[0036] Different bands have different responses to different surface characteristics. For example, the green band reflects more strongly on green leaves of vegetation, the blue band reflects more strongly on water bodies, and the red band reflects more strongly on exposed surfaces.

[0037] By analyzing the reflectance data of each band, different surface cover types, such as vegetation, water bodies, bare soil, etc., can be more accurately identified.

[0038] The acquired reflectivity data can be used to identify different surface characteristics, such as vegetation, water bodies, soil, etc., and the remote sensing images can be divided into different layers according to the identified surface characteristics; different bands correspond to different surface characteristics and materials. In order to improve the accuracy of the data, multiple bands can be preset, and the interference items of the layered remote sensing images can be filtered according to the preset bands to obtain layered images, including removing factors such as clouds, ice and snow that interfere with the remote sensing images, so as to obtain clearer and more accurate surface information.

[0039] Among them, the preset multiple bands include: G (green band), B (blue band), R (red band), NIR (near infrared), MIR (mid infrared), SWIR1 (first short-wave infrared band) and SWIR2 (first short-wave infrared band), each band has its specific wavelength range, the wavelength range of the green band is 0.492-0.577μm, the wavelength range of the blue band is 0.44-0.475μm, the wavelength range of the red band is 0.625-0.74μm, the wavelength range of the near infrared is 0.7-2.5μm, the wavelength range of the mid infrared is 2.5-25μm, the wavelength range of the first short-wave infrared band is 1.56-1.66μm and the wavelength range of the second short-wave infrared band is 2.1-2.3μm. By selecting and using specific bands, interference items not related to the above bands can be removed in a targeted manner. By removing interference factors such as clouds, ice and snow, subsequent land cover classification, change detection and other analyses can be more accurate. Filtering also helps improve the clarity and readability of remote sensing images, making it easier to observe and interpret surface information.

[0040] By performing layered processing and interference filtering on remote sensing images, the quality and accuracy of remote sensing images are improved, providing a more reliable data basis for subsequent analyses such as surface cover classification and environmental monitoring.

[0041] Step S103, obtaining the spatial bottom layer of the monitoring area according to the layered image, wherein the spatial bottom layer includes a wetland vegetation layer.

[0042] Specifically, after the remote sensing images are layered, it is necessary to extract the spatial bottom layer information of the monitoring area from the layered remote sensing images, and the spatial bottom layer includes important information such as the wetland vegetation layer, so as to facilitate more accurate analysis and monitoring of the wetland vegetation in the future.

[0043] Wherein, step S103 includes: determining the spatial distance between the remote sensing image and the satellite based on the reflectivity data; and deducting the spatial distance according to the reflectivity data to obtain the spatial bottom layer of the monitoring area.

[0044] Specifically, since the intensity of reflectivity is related to the distance of the object from the sensor, the reflectivity data can also be used to infer the spatial distance between the remote sensing image and the satellite. By analyzing and processing the reflectivity data in the remote sensing image, the spatial distance between different surface covers and the satellite can be estimated; by setting a certain reflectivity threshold, the pixel points corresponding to the specific surface cover are screened out, thereby removing the information irrelevant to the spatial bottom layer, and the spatial distance is deleted to obtain the spatial bottom layer of the monitoring area. The spatial bottom layer mainly contains the data directly related to the surface in the monitoring area, and eliminates other interference factors such as the atmosphere and clouds, so as to more accurately analyze and understand the actual surface conditions of the monitoring area; it helps to improve the accuracy and pertinence of remote sensing image analysis, and provides a more reliable data basis for subsequent environmental monitoring, surface cover classification and other applications.

[0045] Step S104, performing point cloud coordinate layout on the wetland vegetation layer using a preset point cloud model to generate a data floating point layer.

[0046] Specifically, the preset point cloud model is used to perform point cloud coordinate layout of the wetland vegetation layer, and a three-dimensional coordinate is assigned to each point of the wetland vegetation layer, thereby generating a data floating point layer, which provides a basis for subsequent data comparison and analysis.

[0047] The training process of the point cloud model includes: obtaining historical remote sensing image data sets and performing image recognition to obtain image elements, using rendering model image elements to perform model rendering training to generate a three-dimensional rendering model, and the image elements include mountains, water, lakes, rivers and vegetation; using a three-dimensional rendering model to identify the color and shape of images in the historical remote sensing image data sets to obtain feature information; based on the feature information, marking the key features and contours of the image elements to generate a historical virtual line map; matching the historical virtual line map with the corresponding historical remote sensing image, evaluating the matching success rate, until the matching success rate reaches a preset threshold, and obtaining a trained point cloud model.

[0048] Specifically, a historical remote sensing image data set is obtained, which includes a large number of historical remote sensing images. The historical remote sensing images may be remote sensing images of different times and locations, so as to provide more extensive and diverse training samples.

[0049] Image recognition technology is used to identify historical remote sensing image data, and to identify and separate different image elements in remote sensing images, such as mountains, water, lakes, rivers and vegetation.

[0050] After the image elements are identified, the rendering model is used to train the rendering of each image element, so that the model can generate a three-dimensional rendering model based on the information in the remote sensing image.

[0051] Among them, the rendering model can adopt the ray tracing rendering model, which generates the final image by tracking the reflection, refraction and scattering process of light in the scene; or adopt the rasterization rendering model, which renders the image by drawing pixels on the screen; or adopt the shading rendering model, which is based on the rendering technology of shading programming, and controls the color, texture and lighting effects of the image by writing shading programs, so as to achieve highly customizable graphics rendering. In actual use, you can choose the corresponding rendering model according to different needs to achieve the best graphics effect.

[0052] After the model rendering training is completed, the models corresponding to each image obtained by the rendering training are recognized based on color and shape. For example, a large number of rendered images containing different colors and shapes are collected and annotated for subsequent training and testing. Computer vision technology is used to extract color and shape features in the image, such as color histogram, texture features, shape features, etc.; machine learning algorithms, such as deep learning and support vector machines, are used to establish color and shape recognition models, and historical data sets are used for model training. The trained model is evaluated and optimized according to the evaluation results to improve its recognition accuracy; the trained model is applied to the rendered image to identify the color and shape and extract the feature information in the model.

[0053] Based on the identified feature information, the corresponding historical virtual line map is generated. The historical virtual line map marks the key features and contours of each image element in the historical remote sensing image in the form of a two-dimensional image.

[0054] Each historical remote sensing image and the corresponding historical virtual line map are stored in the same unit in the point cloud model to ensure the correspondence between the historical remote sensing image and the historical virtual line map, which is convenient for subsequent matching operations.

[0055] In order to train the model to extract feature information corresponding to the virtual line map from the remote sensing image, historical remote sensing images and historical virtual line maps are matched with each other. During the matching process, the matching success rate is continuously evaluated. When the matching success rate reaches the preset threshold, it is determined that the point cloud model has sufficient recognition and rendering capabilities, and the training of the point cloud model is completed.

[0056] Through the above training process, the point cloud model can identify and render different geographic features from remote sensing images, providing an accurate basic model for the subsequent point cloud coordinate layout and data floating-point layer generation.

[0057] Among them, step S104 includes: inputting the wetland vegetation layer into a preset point cloud model, and outputting a rendering modeling map, which includes a number of image elements; gridding the rendering modeling map to obtain a virtual line map, and forming point cloud coordinates corresponding to all image elements on the virtual line map based on reflectivity data; creating floating points corresponding to the point cloud coordinates through the image frame number of the wetland vegetation layer, combining with the reverse line map to generate a data floating point layer, and the reverse line map is a line map with the opposite order to the virtual line map.

[0058] Specifically, the remote sensing image of the wetland vegetation layer in the monitoring area is obtained using the preset point cloud model. The point cloud model is a three-dimensional model that can represent the point cloud data of the surface of an object, that is, the three-dimensional coordinates of a series of points on the surface of the object. According to the information in the wetland vegetation layer, the output is a rendered modeling map, which includes various geographical features in the monitoring area, such as mountains, water, lakes, rivers, vegetation and other image elements.

[0059] The rendered modeling image is gridded and divided into a series of small grid units, each of which represents a specific geographical area. Based on the gridding process, a virtual line map is generated, in which lines are used to mark different geographical features in the rendered modeling image, such as the outline of mountains, the boundary of water bodies, the distribution of vegetation, etc. The reflectivity data is used to form corresponding point cloud coordinates for each geographical feature on the virtual line map, and the point cloud coordinates represent the specific position of each geographical feature in three-dimensional space.

[0060] According to the number of image frames of the remote sensing image, a corresponding floating point is created for each point cloud coordinate. The floating point is a discrete point representing each geographic feature in three-dimensional space. According to the virtual line map, a reverse line map with the opposite order is obtained. The reverse line map and the floating point are combined to obtain a data floating point layer. The data floating point layer contains a three-dimensional data set of all geographic feature point cloud coordinates, which can be used for subsequent spatial analysis and visualization.

[0061] Through the above steps, the three-dimensional spatial positions of various geographical features in the monitoring area can be accurately obtained and represented, providing an important data basis for subsequent environmental monitoring, geospatial analysis, etc.

[0062] Step S105, according to the monitoring cycle, the cloud coordinates of the data floating point layer are compared with the historical data floating point layer of the previous cycle to obtain differentiated values, and the differentiated values ​​are evaluated to obtain the wetland vegetation degradation assessment result.

[0063] Specifically, the wetland vegetation degradation monitoring is set with a corresponding monitoring cycle, for example, twice a year. Data comparison is performed regularly according to the monitoring cycle, and the cloud coordinates of the current data floating-point layer are compared with the historical data floating-point layer of the previous cycle to obtain differentiated values, which are used to evaluate the degradation tendency of wetland vegetation. Regular monitoring and evaluation help to timely discover the degradation of wetland vegetation and take corresponding protective measures.

[0064] Among them, step S105 includes: according to the monitoring cycle, obtaining the historical data floating-point layer of the previous cycle, and placing it in the same coordinate system as the current data floating-point layer; identifying the coordinate positions of each floating point in the historical data floating-point layer and the data floating-point layer, and classifying the coordinate positions based on the image elements to obtain classification results; matching the floating points in the historical data floating-point layer and the data floating-point layer to obtain matching floating-point pairs; forming a connecting vector between two floating points based on the matching floating-point pairs, and judging whether the direction of the connecting vector is forward or reverse based on the classification results, and obtaining a differentiated value based on the connecting vector and its direction; based on the differentiated values ​​of all floating-point pairs, evaluating the degradation tendency of wetland vegetation to obtain a wetland vegetation degradation evaluation result.

[0065] Specifically, according to the monitoring cycle, the historical data floating point layer of the previous cycle is obtained, and it is placed in a coordinate system with the current data floating point layer to facilitate data comparison at two time points in the same spatial reference system.

[0066] In the coordinate system, the coordinate positions of each floating point in the historical data floating point layer and the current data floating point layer are identified, and all coordinate positions are classified according to the image elements, including mountains, water, lakes, rivers, vegetation and other categories, to obtain classification results, so as to facilitate targeted degradation tendency analysis according to different geographical features.

[0067] Match each floating point in the historical data floating point layer and the current data floating point layer, find the corresponding geographic feature points that exist at both time points, and obtain matching floating point pairs.

[0068] After the matching is completed, the change difference between the two floating points is determined based on the floating point pair, and the connecting vector between the two floating points is formed through the floating point pair, that is, the direction vector from the historical floating point to the current floating point. According to the coordinate position and time, the direction of the connecting vector is determined, which represents the change trend of the geographical feature points between different time points. According to the classification results of the floating points, the connecting vector is identified as positive or negative, and the differentiated value is obtained according to the connecting vector and its direction; based on the differentiated values ​​of all floating point pairs, the degradation tendency of wetland vegetation is evaluated, and the growth or degradation trend of wetland vegetation is represented by positive and reverse directions; if it is reverse, it means that the vegetation is reduced or the quality is reduced, and it is determined that the wetland vegetation has a tendency to degrade.

[0069] Through the above steps, the changes in wetland vegetation can be monitored regularly, the monitoring process can be automated, the monitoring efficiency can be improved, and degradation trends can be discovered in time so that corresponding protective measures can be taken.

[0070] In this embodiment, by acquiring a remote sensing image of the monitoring area, the reflectivity data of the monitoring area is calculated, the remote sensing image is layered based on the reflectivity data, and interference items such as cloud layers, ice layers and snow layers are filtered to obtain a layered image, thereby ensuring the accuracy and reliability of the detection data, which is suitable for wetland vegetation monitoring in severe weather; the spatial bottom layer of the monitoring area is obtained according to the layered image, and the spatial bottom layer includes a wetland vegetation layer; the wetland vegetation layer is laid out with point cloud coordinates through a preset point cloud model to generate a data floating-point layer, and according to the monitoring period, the data floating-point layer is compared with the historical data floating-point layer of the previous period for cloud coordinates to obtain differentiated values, and the differentiated values ​​are evaluated to obtain a wetland vegetation degradation assessment result, thereby realizing high-precision automated monitoring of wetland vegetation, improving monitoring efficiency, and being able to periodically assess the tendency of wetland vegetation degradation, which is helpful to timely discover wetland vegetation degradation trends and provide a scientific basis for wetland protection.

[0071] like Figure 2 As shown, a wetland vegetation degradation monitoring device 20 based on satellite remote sensing images is provided, which is used to implement a wetland vegetation degradation monitoring method based on satellite remote sensing images as described above, including: a remote sensing image acquisition module 21, a remote sensing image layering module 22, a spatial bottom layer acquisition module 23, a point cloud coordinate layout module 24 and a vegetation degradation assessment module 25, wherein:

[0072] The remote sensing image acquisition module 21 is used to acquire the remote sensing image of the monitoring area and calculate the reflectivity data of the monitoring area according to the remote sensing image;

[0073] A remote sensing image layering module 22 is used to perform layering processing on the remote sensing image based on the reflectivity data and filter interference items to obtain a layered image, wherein the interference items include cloud layers, ice layers and snow layers;

[0074] The spatial bottom layer acquisition module 23 is used to obtain the spatial bottom layer of the monitoring area according to the layered image, and the spatial bottom layer includes a wetland vegetation layer;

[0075] The point cloud coordinate layout module 24 is used to perform point cloud coordinate layout on the wetland vegetation layer through a preset point cloud model to generate a data floating point layer;

[0076] The vegetation degradation assessment module 25 is used to compare the cloud coordinates of the data floating point layer with the historical data floating point layer of the previous period according to the monitoring period to obtain the differentiated values, and evaluate the differentiated values ​​to obtain the wetland vegetation degradation assessment result.

[0077] In one embodiment, the remote sensing image acquisition module 21 is specifically used to: receive the location information of the monitoring area and synchronize it to the satellite remote sensing system; obtain the remote sensing image corresponding to the location information through the satellite remote sensing system, and obtain the resolution of the monitoring area; obtain various reflectivities in the remote sensing image at the same time, and calculate the various reflectivities to obtain reflectivity data.

[0078] In one embodiment, the remote sensing image layering module 22 is specifically used to: identify different surface characteristics based on reflectivity data, and layer the remote sensing image according to the surface characteristics; obtain preset bands, and filter interference items on the layered remote sensing image according to the preset bands to obtain a layered image, wherein the preset bands include a green band, a blue band, a red band, a near-infrared band, a mid-infrared band, a first short-wave infrared band, and a second short-wave infrared band.

[0079] In one embodiment, the spatial bottom layer acquisition module 23 is specifically used to: determine the spatial distance between the remote sensing image and the satellite based on the reflectivity data; and reduce the spatial distance according to the reflectivity data to obtain the spatial bottom layer of the monitoring area.

[0080] In one embodiment, the point cloud coordinate layout module 24 is specifically used to: input the wetland vegetation layer into a preset point cloud model, and output a rendered modeling diagram, which includes a number of image elements; grid the rendered modeling diagram to obtain a virtual line map, and form point cloud coordinates corresponding to all image elements on the virtual line map based on reflectivity data; create floating points corresponding to the point cloud coordinates through the image frame number of the wetland vegetation layer, and combine the reverse line map to generate a data floating point layer, and the reverse line map is a line map with the opposite order to the virtual line map.

[0081] In one embodiment, the vegetation degradation assessment module 25 is specifically used to: obtain the historical data floating-point layer of the previous period according to the monitoring period, and place it in the same coordinate system as the current data floating-point layer; identify the coordinate positions of each floating point in the historical data floating-point layer and the data floating-point layer, classify the coordinate positions based on the image elements, and obtain classification results; match the floating points in the historical data floating-point layer and the data floating-point layer to obtain matching floating-point pairs; form a connecting vector between two floating points based on the matching floating-point pairs, and judge whether the direction of the connecting vector is forward or reverse based on the classification results, and obtain a differentiated value based on the connecting vector and its direction; based on the differentiated values ​​of all floating-point pairs, evaluate the degradation tendency of wetland vegetation to obtain a wetland vegetation degradation assessment result.

[0082] In one embodiment, an electronic device is provided. The device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the device is used to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the device is used to store configuration templates and can also be used to store target web page data. The network interface of the device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a wetland vegetation degradation monitoring method based on satellite remote sensing images is implemented.

[0083] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the device to which the scheme of the present application is applied. The specific device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0084] In one embodiment, a computer-readable storage medium may also be provided, wherein the storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the method described in the aforementioned embodiment. The computer may be part of the above-mentioned wetland vegetation degradation monitoring device based on satellite remote sensing images.

[0085] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0086] Obviously, those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a computer storage medium (ROM / RAM, magnetic disk, optical disk) and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than that here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Therefore, the present invention is not limited to any specific combination of hardware and software.

[0087] The above contents are further detailed descriptions of the present invention in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A method for monitoring wetland vegetation degradation based on satellite remote sensing images, characterized in that: The following steps are involved: Acquire a remote sensing image of the monitoring area, and calculate reflectivity data of the monitoring area based on the remote sensing image; Performing layered processing on the remote sensing image based on the reflectivity data and filtering interference items to obtain a layered image, wherein the interference items include cloud layers, ice layers, and snow layers; According to the layered image, a spatial bottom layer of the monitoring area is obtained, wherein the spatial bottom layer includes a wetland vegetation layer; The point cloud coordinates of the wetland vegetation layer are laid out through a preset point cloud model to generate a data floating point layer, including: inputting the wetland vegetation layer into a preset point cloud model, and outputting a rendering modeling map, wherein the rendering modeling map includes a number of image elements; gridding the rendering modeling map to obtain a virtual line map, wherein the virtual line map uses lines to mark different geographical features in the rendering modeling map; forming point cloud coordinates corresponding to all image elements on the virtual line map based on the reflectivity data; creating floating points corresponding to the point cloud coordinates through the image frame number of the wetland vegetation layer, and combining the reverse line map to generate a data floating point layer, wherein the reverse line map is a line map with a reverse order to the virtual line map; The training process of the point cloud model includes: obtaining a historical remote sensing image data set, performing image recognition on the images in the historical remote sensing image data set to obtain image elements, using a rendering model to perform model rendering training on the image elements to generate a three-dimensional rendering model, wherein the image elements include mountains, water, lakes, rivers and vegetation; using the three-dimensional rendering model to perform color and shape recognition on the images in the historical remote sensing image data set to obtain feature information; marking the key features and contours of the image elements according to the feature information to generate a historical virtual line map; matching the historical virtual line map with the corresponding historical remote sensing image, evaluating the matching success rate, until the matching success rate reaches a preset threshold, and obtaining a trained point cloud model; According to the monitoring cycle, the cloud coordinates of the data floating point layer are compared with the historical data floating point layer of the previous cycle to obtain differentiated values, and the differentiated values ​​are evaluated to obtain the wetland vegetation degradation assessment result.

2. The wetland vegetation degradation monitoring method based on satellite remote sensing images according to claim 1 is characterized in that: The step of acquiring a remote sensing image of the monitoring area and calculating reflectivity data of the monitoring area according to the remote sensing image includes: Receive monitoring area location information and synchronize it to the satellite remote sensing system; Acquire a remote sensing image corresponding to the location information through a satellite remote sensing system, and obtain the resolution of the monitoring area; At the same time, each reflectivity in the remote sensing image is obtained, and each reflectivity is calculated to obtain reflectivity data.

3. The wetland vegetation degradation monitoring method based on satellite remote sensing images according to claim 1 is characterized in that: The step of performing layered processing on the remote sensing image based on the reflectivity data and filtering interference items to obtain a layered image includes: Performing layered processing on the remote sensing image based on the reflectivity data; Acquire preset bands, and perform interference filtering on the layered remote sensing image according to the preset bands to obtain a layered image, wherein the preset bands include a green band, a blue band, a red band, a near infrared band, a mid-infrared band, a first short-wave infrared band, and a second short-wave infrared band.

4. The wetland vegetation degradation monitoring method based on satellite remote sensing images according to claim 1 is characterized in that: The step of obtaining the spatial bottom layer of the monitoring area according to the layered image comprises: Based on the reflectivity data, determining the spatial distance between the remote sensing image and the satellite; The spatial distance is subtracted according to the reflectivity data to obtain the spatial bottom layer of the monitoring area.

5. The wetland vegetation degradation monitoring method based on satellite remote sensing images according to claim 1 is characterized in that: According to the monitoring cycle, the data floating point layer is compared with the historical data floating point layer of the previous cycle in cloud coordinates to obtain a differentiated value, and the differentiated value is evaluated to obtain a wetland vegetation degradation assessment result, including: According to the monitoring cycle, obtain the historical data floating point layer of the previous cycle and place it in the same coordinate system as the current data floating point layer; Identify the coordinate positions of each floating point in the historical data floating point layer and the data floating point layer, classify the coordinate positions based on the image elements, and obtain classification results; Matching the floating points in the historical data floating point layer and the data floating point layer to obtain matching floating point pairs; Forming a connection vector between two floating points according to the matched floating point pairs, and judging whether the direction of the connection vector is forward or reverse based on the classification result, and obtaining a differentiated value according to the connection vector and its direction; Based on the differentiated values ​​of all floating point pairs, the degradation tendency of wetland vegetation is evaluated and the wetland vegetation degradation assessment results are obtained.

6. A wetland vegetation degradation monitoring device based on satellite remote sensing images, characterized in that: A method for monitoring wetland vegetation degradation based on satellite remote sensing images as claimed in any one of claims 1 to 5, comprising: A remote sensing image acquisition module is used to acquire a remote sensing image of the monitoring area and calculate the reflectivity data of the monitoring area based on the remote sensing image; A remote sensing image layering module, used for performing layering processing on the remote sensing image based on the reflectivity data and filtering interference items to obtain a layered image, wherein the interference items include cloud layers, ice layers and snow layers; A spatial bottom layer acquisition module, used to obtain the spatial bottom layer of the monitoring area according to the layered image, wherein the spatial bottom layer includes a wetland vegetation layer; A point cloud coordinate layout module is used to perform point cloud coordinate layout on the wetland vegetation layer through a preset point cloud model to generate a data floating point layer, including: inputting the wetland vegetation layer into a preset point cloud model, and outputting a rendering modeling diagram, wherein the rendering modeling diagram includes a number of image elements; gridding the rendering modeling diagram to obtain a virtual line map, wherein the virtual line map uses lines to mark different geographical features in the rendering modeling diagram; forming point cloud coordinates corresponding to all image elements on the virtual line map based on the reflectivity data; creating floating points corresponding to the point cloud coordinates through the image frame number of the wetland vegetation layer, and generating a data floating point layer in combination with a reverse line map, wherein the reverse line map is a line map in the opposite order of the virtual line map; The training process of the point cloud model includes: obtaining a historical remote sensing image data set, performing image recognition on the images in the historical remote sensing image data set to obtain image elements, using a rendering model to perform model rendering training on the image elements to generate a three-dimensional rendering model, wherein the image elements include mountains, water, lakes, rivers and vegetation; using the three-dimensional rendering model to perform color and shape recognition on the images in the historical remote sensing image data set to obtain feature information; marking the key features and contours of the image elements according to the feature information to generate a historical virtual line map; matching the historical virtual line map with the corresponding historical remote sensing image, evaluating the matching success rate, until the matching success rate reaches a preset threshold, and obtaining a trained point cloud model; The vegetation degradation assessment module is used to compare the cloud coordinates of the data floating point layer with the historical data floating point layer of the previous period according to the monitoring period to obtain differentiated values, and evaluate the differentiated values ​​to obtain the wetland vegetation degradation assessment result.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images