Method of, system for, equipment for, and medium for detecting forest land disturbance based on full polarization SAR image

Abstract

To provide a method of, a system for, electronic equipment for, and a storage medium for detecting forest land disturbance as a better service by efficient, precise and reliable forest land resource monitoring, environment protection, and ecological protection restoration supervision management through forest land disturbance detection based on a full polarization SAR image.SOLUTION: A method of detecting forest land disturbance based on a full polarization SAR image has the steps of acquiring a full polarization SAR image in a target region, segmenting the image into that of a reference time phase and that of a monitoring time phase, subjecting the segmented full polarization SAR image to pre-processing including a polarization matrix conversion of the SAR image, Refined Lee filtering, and false color synthesis, acquiring a false color synthesis SAR image, inputting the false color synthesis SAR image into an optimized ground surface change detection model to extract a ground surface change patch, combining the patch with a radar vegetation index RVI image to acquire an initial forest land disturbance patch, carrying out overlapping check for the initial forest land disturbance patch to acquire a forest land disturbance patch in the target region, and generating a visualized result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the field of remote sensing technology, and more particularly to a method, system, apparatus and medium for forest disturbance detection based on fully polarimetric SAR images. [Background technology]

[0002] In recent years, our country has attached great importance to ecological environment protection and ecological civilization construction, implemented a series of effective ecological environment protection measures, and made remarkable improvements in the quality of the ecological environment.

[0003] Forests have functions such as water and soil conservation, water source nutrition, windbreaks and sand fixation, and biodiversity protection, and are an important component of the global ecosystem and ecological balance. Particularly in areas with important ecological functions and fragile ecological environments, excessive forest development is a major factor in the decline of regional ecological service functions. In recent years, cases of illegal destruction of forest resources have often persisted despite being prohibited, posing a serious threat to national ecological security. Therefore, strengthening monitoring of forest development is urgent and extremely important for enhancing ecosystem diversity, stability, and sustainability.

[0004] Remote sensing technology has been widely applied in the field of land cover change monitoring. Currently, land surface change detection technology based on high-resolution optical remote sensing imagery has become the primary tool for forest land regeneration and other tasks. However, optical imagery is heavily affected by weather, and in areas with high forest cover and heavy cloud and rain, optical data acquisition is often difficult or the acquired image quality is poor, which to some extent affects the accuracy and timeliness of remote sensing monitoring data results.

[0005] Compared with optical satellites, the number of SAR satellites is relatively small, and the data source limits its widespread application to some extent. At the same time, SAR images are affected by factors such as speckle noise, geometric distortion, layover, and shadows, making them more difficult to use in monitoring human activities, and there has been relatively little related application research. In recent years, SAR satellite technology has developed rapidly, with multiple high-resolution satellites of different bands launched, and polarization methods have evolved from single polarization to dual polarization, multi-polarization, and full polarization, further expanding the application fields and directions of SAR satellite remote sensing.

[0006] Therefore, how to design a forest disturbance detection method based on full-polarization SAR images, compensate for the lack of coverage of high-resolution optical images in areas with heavy clouds and rain, and optimize the monitoring of forest development activities is a problem that needs to be solved urgently by those skilled in the art. Summary of the Invention [Problem to be solved by the invention]

[0007] In view of this, the present invention provides a forest disturbance detection method based on full-polarization SAR images, which can be applied to forest disturbance monitoring in important ecological spaces, so as to detect forest disturbance activities in a timely manner, serve ecological protection and restoration supervision and management, and further improve the diversity, stability, and sustainability of ecosystems. [Means for solving the problem]

[0008] To achieve the above objectives, the present invention adopts the following technical solutions:

[0009] According to a first aspect, the present invention provides a method for forest disturbance detection based on fully polarimetric SAR images, said method comprising: S1: acquiring a full-polarization SAR image within a target area, and dividing the full-polarization SAR image into a reference time phase and a monitoring time phase; S2. Perform preprocessing on the segmented full-polarization SAR image, including SAR image polarization matrix transformation, refined Lee filtering, and false color synthesis, to obtain a false color synthesis SAR image; S3. Input the false color synthetic SAR image into an optimized ground change detection model to extract ground change patches, and combine them with radar vegetation index (RVI) images to obtain initial forest disturbance patches; S4: performing an overlay check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generate a visualization result.

[0010] Here, the step S2 of performing preprocessing on the divided full-polarization SAR images to obtain a false color synthetic SAR image specifically includes the following steps:

[0011] S21. SAR image input: converting the full polarimetric SAR images of the reference time phase and the monitoring time phase into a standard data format and performing radiation correction processing to obtain full polarimetric single-view composite data X; X={X i :i=1,2,…,M} where X i ={xpq i =Ipq i +jQpq i :p,q∈{H,V}} M is the total number of pixels, p is the antenna transmission method, q is the antenna reception method, i is the pixel index, j is the imaginary unit, xpq i is the complex scattering coefficient of the ith pixel in the pq polarization scheme, Ipq i xpq i The real part of Qpq i xpq i where H is the imaginary part of the horizontal polarization and V is the vertical polarization.

[0012] S22, polarized matrix transformation of SAR image: constructing a polarized scattering matrix S based on the full polarized single-view composite data X; S={S i :i=1,2,…,M} TIFF2025170779000002.tif16166The polarization scattering matrix S is used to define a target scattering vector set K, K={K i :i=1,2,…,M} TIFF2025170779000003.tif16166 Convert the target scattering vector set K into a polarization coherence matrix T, T={T i :i=1,2,…,M} TIFF2025170779000004.tif23166 where * indicates conjugate operation, < > indicates multi-view processing,

[0013] S23, Refined Lee filtering: performing refined Lee filtering on the polarization coherent matrix T to obtain a filtered polarization coherent matrix T1, including: TIFF2025170779000005.tif20166 where N is the number of views, m is the pixel index of the image after multi-view processing, If we change the azimuth resolution to 1 / N of the original and leave the distance resolution unchanged, You can get TIFF2025170779000006.tif10166, where: TIFF2025170779000007.tif10166 denotes the average value of the same boundary matching window within the local window in which the m-th pixel is located, and w is the filtering weight value;

[0014] S24, false color synthesis: full polarization false color synthesis is performed on the filtered polarization transformation matrix T1 using the Pauli synthesis method to obtain a false color synthesis SAR image S1 and a false color synthesis SAR image S2.

[0015] Furthermore, in S3, the optimization of the ground surface change detection model is performed by: S31, creating a tag image data set based on the surface disturbance interpretation flag; S32, training a ground change detection model using the tag image dataset; S33: classifying the false color synthetic SAR image using a trained ground change detection model to obtain a classification result; S34, performing vector transformation on the classification result, calculating a cross-entropy loss value using a cross-entropy loss function, and performing model evaluation; S35, if the cross-entropy loss value is less than a cross-entropy loss threshold, output the classification result; otherwise, return to S31 and use the classification result to supplement the tag image dataset.

[0016] Furthermore, in the step S31, the construction of the ground disturbance interpretation flag is performed as follows: S311, acquiring two high-resolution optical remote sensing images, one at an early stage and one at a late stage, corresponding to the time phase of the fully polarized SAR image; S312, inputting the two high-resolution optical remote sensing images from the early and late periods into a pre-training neural network model to extract newly added ground disturbance patches; S313, combining the false color synthetic SAR image and constructing a ground disturbance interpretation flag.

[0017] Furthermore, in S3, the Freeman method is adopted to calculate the radar vegetation index (RVI); RVI=Fv / (Fv+Fd+Fs) Here, Fv, Fd, and Fs represent the odd scattering component, the even scattering component, and the volume scattering component obtained by decomposing the scattering matrix after polarization matrix transformation of the SAR image, respectively.

[0018] Furthermore, in S4, performing an overlap check on the initial forest disturbance patch includes: S411, performing a registration check on the initial forest land disturbance patch and forestry land data to obtain a forestry land disturbance patch; S412, performing a spatial registration check on the forestry land disturbance patches and the slope data to obtain forestry land disturbance patches within a preset slope range.

[0019] Furthermore, in S4, generating the visualization result includes: S421, deriving forest disturbance patches within the target area from a vector data layer; S422, combining with existing attribute information and marking the vector data layer; S423, generating multiple types of visualization results, including a forest land property classification map, a forest land disturbance heat map, and a forest land disturbance trend graph, based on the marked vector data layer.

[0020] According to a second aspect, the present invention provides a forest disturbance detection system based on fully polarimetric SAR imagery, said system comprising: a data acquisition module for acquiring a full-polarization SAR image within a target area and for segmenting the full-polarization SAR image into a reference time phase and a monitoring time phase; a data preprocessing module for performing preprocessing on the segmented full-polarization SAR image, including SAR image introduction, SAR image polarization matrix transformation, refined Lee filtering, and false color synthesis, to obtain a false color synthesis SAR image; a feature extraction module for inputting the false-color synthetic SAR image into an optimized ground change detection model to extract ground change patches, and combining the false-color synthetic SAR image with a radar vegetation index (RVI) image to obtain initial forest disturbance patches; and a registration check module for performing a registration check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generate a visualization result.

[0021] According to a third aspect, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and operable to run on the processor, the above-mentioned forest disturbance detection method being realized when the processor executes the computer program.

[0022] According to a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the forest disturbance detection method described above.

[0023] For the explanation of the second to fourth aspects of the present invention, the detailed explanation of the first aspect may be referred to, and the beneficial effects of the explanation of the second to fourth aspects can be analyzed by referring to the beneficial effects of the first aspect, so they will not be explained here. [Effects of the Invention]

[0024] As can be seen from the above technical solutions, the forest disturbance detection method based on full polarimetric SAR images according to the present invention has the following beneficial effects compared with the prior art:

[0025] 1. This method utilizes full-polarization SAR images and combines them with image segments from the reference and monitoring phases to improve the accuracy of detecting and analyzing forest disturbances within the target area. Full-polarization SAR images have weather-independent and remote sensing transparency, and can obtain high-quality image data under various weather conditions.

[0026] 2. For full-polarization SAR images, preprocessing including SAR image introduction, SAR image polarization matrix transformation, refined Lee filtering, and false color synthesis is performed to generate high-quality false color synthetic SAR images, which can provide more intuitive and clearer ground surface information, and is advantageous for subsequent ground change detection and patch extraction.

[0027] 3. Based on the optimized ground change detection model, ground change patches can be extracted more accurately. In addition, by combining with radar vegetation index (RVI) images, disturbance situations in forest areas can be better distinguished, and the accuracy and reliability of detection results can be improved.

[0028] 4. Finally, by performing an overlay check on the initial forest disturbance patches, the accuracy of forest disturbance detection within the target area can be further improved, reducing the false positive and false negative rates and producing more reliable and accurate visualization results, providing important reference information for forest protection and management. [Brief explanation of the drawings]

[0029] In order to more clearly describe the technical solutions in the embodiments of the present invention or the existing technology, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the existing technology. It is obvious that the accompanying drawings in the following description are only the embodiments of the present invention, and those skilled in the art can obtain other accompanying drawings based on the accompanying drawings provided without any creative efforts. [Figure 1] 2 is a flowchart of a method for detecting forest disturbance based on fully polarimetric SAR images according to an embodiment of the present invention. [Figure 2] 1 is a method flowchart for performing pre-processing on a segmented fully polarimetric SAR image according to an embodiment of the present invention. [Figure 3] 1 is a schematic diagram of a full-polarization single-view composite data image after the introduction of SAR imaging according to an embodiment of the present invention; FIG. [Figure 4] FIG. 10 is a schematic diagram of an image after polarization matrix transformation according to an embodiment of the present invention. [Figure 5] 1 is a schematic diagram of an image subjected to Refined Lee filtering according to an embodiment of the present invention; [Figure 6] 1 is a diagram illustrating an SAR image after false color synthesis according to an embodiment of the present invention. [Figure 7] 1 is a flowchart of a method for optimizing a ground change detection model according to an embodiment of the present invention. [Figure 8] 4 is a flowchart of a method for constructing a ground disturbance interpretation flag according to an embodiment of the present invention. [Figure 9] FIG. 1 is a schematic diagram of a newly added ground disturbance patch according to an embodiment of the present invention. [Figure 10] FIG. 10 is a schematic diagram of a ground disturbance interpretation flag according to an embodiment of the present invention. [Figure 11] FIG. 2 is a radar vegetation index image diagram according to an embodiment of the present invention. [Figure 12] FIG. 1 is a schematic diagram of an initial forest disturbance patch according to an embodiment of the present invention. [Figure 13] 1 is a flowchart of a method for performing an overlay check on an initial forest disturbance patch according to an embodiment of the present invention. [Figure 14] 1 is a flowchart of a method for generating a visualization result according to an embodiment of the present invention. [Figure 15] 1 is a structural schematic diagram of a forest disturbance detection system based on a fully polarized SAR image according to an embodiment of the present invention; [Figure 16] 1 is a structural schematic diagram of an electronic device according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0030] The following clearly and completely describes the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without any creative effort are all within the scope of protection of the present invention.

[0031] The detection method according to the embodiment of the present application can be applied to a detection server. The detection server may be hardware or software. If the detection server is hardware, it may be implemented as a distributed server cluster providing detection services, or as a single server. If the detection server is software, it may be installed on the servers listed above. It may be implemented as multiple software or software modules, or as a single software or software module, and is not particularly limited here.

[0032] Example 1: As shown in FIG. 1 , this embodiment provides a forest disturbance detection method based on full polarimetric SAR images, taking an area within a certain range of a city as a target area, and the method includes:

[0033] S1, acquiring a full-polarization SAR image within a target area, and dividing the full-polarization SAR image into a reference time phase and a monitoring time phase;

[0034] The domestically produced Gaofen-3 satellite has a variety of polarization modes, including single polarization, dual polarization, and full polarization. It uses Gaofen-3 full polarization data to cover a certain area of ​​a city, with a spatial resolution of 8 meters. To improve identification accuracy, an interpretation flag library has been established, and Gaofen-1 optical data is used as auxiliary data, with a spatial resolution of 2 meters.

[0035] S2. Perform preprocessing including polarization matrix transformation of SAR image, refined Lee filtering and false color synthesis on the segmented full polarization SAR image to obtain a false color synthesis SAR image.

[0036] S3. Input the false color synthetic SAR image into an optimized ground change detection model to extract ground change patches, and combine them with the radar vegetation index (RVI) image to obtain initial forest disturbance patches;

[0037] S4: Perform a registration check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generate a visualization result.

[0038] The method has efficient, accurate and reliable detection capabilities, and can perform ground observations throughout the day, in all weather conditions and through clouds and fog. It can be widely applied to forest resource monitoring, environmental protection and other fields, especially in remote sensing monitoring of forest development activities in areas with heavy clouds and rain where effective optical imaging coverage is lacking. It can better serve the supervision and management of ecological protection and restoration, and can enhance the diversity, stability and sustainability of ecosystems.

[0039] The above steps are explained in more detail below:

[0040] As shown in FIG. 2, in this embodiment S2, preprocessing is performed on the segmented full-polarization SAR image to obtain a false color synthetic SAR image, specifically including:

[0041] S21. SAR image input: converting the full polarimetric SAR images of the reference time phase and the monitoring time phase into a standard data format and performing radiation correction processing to obtain full polarimetric single-view composite data X; X={X i :i=1,2,…,M} where X i ={xpq i =Ipq i +jQpq i :p,q∈{H,V}} M is the total number of pixels, p is the antenna transmission method, q is the antenna reception method, i is the pixel index, j is the imaginary unit, xpq i is the complex scattering coefficient of the ith pixel in the pq polarization scheme, Ipq i xpq i The real part of Qpq i xpq i where H is the imaginary part of the horizontal polarization and V is the vertical polarization.

[0042] As shown in Figure 3, single-view composite data without post-processing is noisy and has poor image readability, making it impossible to directly use for human activity patch interpretation.

[0043] S22, polarization matrix transformation of SAR image: construct a polarization scattering matrix S based on the full polarization single-view composite data X: S={S i :i=1,2,…,M} TIFF2025170779000008.tif16166Define the target scattering vector set K using the polarization scattering matrix S: K={K i :i=1,2,…,M} TIFF2025170779000009.tif16166Transform the target scattering vector set K into a polarization coherence matrix T: T={T i :i=1,2,…,M} TIFF2025170779000010.tif26166 where * indicates conjugate operation, < > indicates multi-view processing,

[0044] As shown in Figure 4, the transformed polarization coherence matrix contains nine images.

[0045] S23. Refined Lee filtering: performing refined Lee filtering on the polarization coherent matrix T, adopting a window size of 7*7 refined Lee filtering, to obtain a filtered polarization coherent matrix T1, including: TIFF2025170779000011.tif20166 where N is the number of views, m is the pixel index of the image after multi-view processing, By changing the azimuth resolution to 1 / N of the original and leaving the range resolution unchanged, we can achieve the following: TIFF2025170779000012.tif10166, where σ denotes the average value of the same boundary matching window within the 7*7 local window in which the Mth pixel is located, w is the filtering weight value,

[0046] As shown in Figure 5, the image after filtering using Refined Lee with a window size of 7*7 has moderate noise and retains the most detailed image information.

[0047] S24, false color synthesis: full polarization false color synthesis is performed on the filtered polarization transformation matrix T1 using the Pauli synthesis method to obtain a false color synthesis SAR image S1 and a false color synthesis SAR image S2.

[0048] As shown in Figure 6, the SAR image after false color synthesis has rich color information, which significantly improves the visibility of different feature images and is advantageous for interpreting human activities.

[0049] As shown in FIG. 7, in this embodiment S3, the optimization of the ground change detection model includes: S31, creating tag image datasets based on the surface disturbance interpretation flags; S32, training a ground change detection model using the tag image dataset; S33, classifying the false color synthetic SAR image using a trained ground change detection model to obtain a classification result; S34, performing vector transformation on the classification result, calculating a cross-entropy loss value using a cross-entropy loss function, and performing model evaluation; S35, if the cross-entropy loss value is less than the cross-entropy loss threshold, output the classification result; otherwise, return to S31 and use the classification result to supplement the tag image dataset.

[0050] As shown in Figure 8, the construction of the surface disturbance interpretation flag includes: S311, acquiring two high-resolution optical remote sensing images, one early and one late, that correspond to the time phase of the fully polarized SAR image; S312, as shown in Figure 9, input the two high-resolution optical remote sensing images from the early and late periods into a pre-training neural network model to extract newly added ground disturbance patches; S313, as shown in FIG. 10, combine with the false color synthetic SAR image to construct a ground disturbance interpretation flag.

[0051] As shown in Figure 11, the Freeman method is adopted to calculate the Radar Vegetation Index RVI: RVI=Fv / (Fv+Fd+Fs) Here, Fv, Fd, and Fs represent the odd scattering component, the even scattering component, and the volume scattering component obtained by decomposing the scattering matrix after polarization matrix transformation of the SAR image, respectively.

[0052] As shown in Figure 12, it is combined with the radar vegetation index (RVI) to obtain the initial forest disturbance patch.

[0053] As shown in FIG. 13 , in this embodiment S4, performing an overlap check on the initial forest disturbance patch includes: S411, performing a registration check on the initial forest disturbance patch and the forestry land data to obtain a forestry land disturbance patch; S412, a spatial overlap check is performed on the forestry land disturbance patch and the gradient data to obtain the forestry land disturbance patch within a preset gradient range.

[0054] Below we provide a detailed description of the registration check performed on the initial forest disturbance patch:

[0055] Performing a registration check on the initial forest disturbance patch in GIS (geographic information system) software, specifically a spatial registration check between forestry land data and slope data:

[0056] Data preparation: Ensure you have a vector data layer containing forestry land data, including forestry land boundaries and related information. Slope data: You also need a vector or raster data layer containing slope information. This can be slope data calculated based on a DEM (Digital Elevation Model). Initial forest disturbance patches: This is the required layer to check and could be a vector layer containing the location and extent of forest disturbances.

[0057] Importing data: Open ArcGIS software and create a new map project. Use the "Add Data" function to import forestry land data, slope data, and initial forest disturbance patches into the map project.

[0058] Spatial Overlay Analysis: Overlay with forestry land data: In the ArcGIS toolbox, find the "Overlay Analysis" tool. Depending on the desired result, choose the "Intersection" or "Discrimination" tool. If you just want to find disturbance patches that overlap with forestry land, you can use the "Intersection" tool; if you want to keep other information of the initial forest disturbance patches, you can use the "Discrimination" tool. In the pop-up dialog box, set the input element (initial forest disturbance patches) and the overlay element (forestry land data).

[0059] Set the location and name of the output layer and click "OK" to run the analysis. The result will be a new layer containing forest disturbance patches that overlap with the forestry land. Overlay with slope data (assuming the slope data is raster data): If the slope data is raster data, you must convert it to vector data using the "Raster to Vector" tool to facilitate overlay analysis. Once converted, use the "Spatial Selection" or "Select by Location" tool to select vector slope data whose slopes fall within a preset range. Then, use the same overlay analysis tool (e.g., "Intersect" or "Distinguish") as for the forestry land data to overlay the selected vector slope data with the forestry land disturbance patches. After setting the parameters and running the analysis, the result will be a new layer containing forestry land disturbance patches within the preset slope range.

[0060] Result handling: Perform any necessary processing or cleanup on the overlay analysis results, such as removing duplicates, merging adjacent patches, etc. Modify layer styles and symbology as needed to better display the results. Use the ArcGIS derive functionality to derive results into the required format, such as Shapefile, GeoJSON, or KMZ, so they can be shared with other software or platforms.

[0061] Through the above steps, an overlay check of forestry land data and slope data is performed on the initial forest land disturbance patch based on ArcGIS software, and a forestry land disturbance patch within the required preset slope range is generated.

[0062] As shown in FIG. 14, generating a visualization result includes: S421, deriving forest disturbance patches within the target area from a vector data layer; S422, combining with existing attribute information and marking the vector data layer; S423, based on the marked vector data layer, generating multiple types of visualization results, including a forest land property classification map, a forest land disturbance heat map, and a forest land disturbance trend graph.

[0063] Forest Disturbance Classification Map: A visualization map that classifies forest disturbance patches according to their nature and represents them with different colors or symbols, allowing users to clearly understand the status of different types of forest disturbances in the target area, such as deforestation, fires, and natural disasters.

[0064] Forest Disturbance Heat Map: Displays forest disturbance heat areas within a target area. Heat areas are typically represented as a high density of points, a thermodynamic or density map, and are used to indicate the concentration and distribution of forest disturbances. This helps determine areas where action needs to be focused.

[0065] Forest disturbance trend graph: Time series data can generate forest disturbance trend graphs. Such visualization results can show the changing trend of forest disturbance over time, helping analysts understand the evolution and influencing factors of forest disturbance.

[0066] The technical solution in this example combines fully polarimetric SAR images with radar vegetation index (RVI), surface change detection model optimization, and the construction of surface disturbance interpretation flags to achieve accurate detection and visualization of forest disturbances, which plays an important role in forest resource management, environmental monitoring, and natural disaster assessment, providing more effective data support and decision-making basis for related fields.

[0067] Example 2: As shown in FIG. 15, this embodiment provides a forest disturbance detection system based on full polarimetric SAR images, which includes: a data acquisition module for acquiring a full-polarization SAR image within a target area and for segmenting the full-polarization SAR image into a reference time phase and a monitoring time phase; a data preprocessing module for performing preprocessing on the segmented full-polarization SAR image, including SAR image introduction, SAR image polarization matrix transformation, refined Lee filtering, and false color synthesis, to obtain a false color synthesis SAR image; a feature extraction module for inputting the false-color synthetic SAR image into an optimized ground change detection model to extract ground change patches, and combining the false-color synthetic SAR image with a radar vegetation index (RVI) image to obtain initial forest disturbance patches; and a registration check module for performing a registration check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generate a visualization result.

[0068] Example 3: As shown in FIG. 16, this embodiment provides an electronic device including a memory, a processor, and a computer program that is stored in the memory and can run on the processor, and when the processor executes the computer program, the disturbance detection method in the above embodiment is realized.

[0069] Example 4: This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the forest disturbance detection method of the above embodiment.

[0070] In the technical solution of the above embodiment, the fully polarized SAR image can collect more detailed information about the ground surface due to different polarization states. Compared with traditional non-polarized SAR images, the fully polarized SAR image can provide additional information about the ground surface structure and material, which is particularly important for forest disturbance detection. Because forest disturbances often cause changes in the ground surface structure, the fully polarized SAR image can capture these subtle structural changes, thereby helping to identify and monitor forest disturbance events.

[0071] SAR image preprocessing includes steps such as image integration, polarization matrix transformation, and refined Lee filtering, which can remove or reduce the effects of noise and improve image quality. False color synthesis can highlight important features while suppressing irrelevant background information, making forest disturbance detection more accurate.

[0072] After the ground change detection model is optimized, it can process preprocessed SAR images more efficiently and accurately extract ground change patches. The overlay check process can determine the specific extent and extent of forest disturbance by comparing SAR images from different time periods. This reduces the error in a single measurement and ensures the accuracy of the final disturbance patch. Converting complex SAR image data into easy-to-understand visualization results provides decision makers and stakeholders with fast and accurate information, helping them respond in a timely manner and take appropriate protective measures.

[0073] By effectively utilizing full-polarization SAR imagery, applying preprocessing and false color synthesis, optimizing the ground change detection model, overlay checking, and generating visualization results, this method has relatively high accuracy and reliability in practical applications, and can effectively support forest management and protection work.

[0074] It should be understood that the methods, systems, devices, and media described in the above embodiments of the present application may be realized in other ways. The above-described embodiments of the methods, systems, devices, and media are merely illustrative. For example, the division into modules or units is merely a logical functional division, and other division methods may be used in actual implementation. Each functional unit may be integrated into one processing unit, each unit may exist physically independently, or two or more units may be integrated into one unit.

[0075] The various exemplary units and algorithm steps described in the embodiments disclosed herein can be combined and realized in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may implement the described functions using different methods for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0076] Here, a computer program includes computer program code, which may be in source code form, object code form, an executable file, or any intermediate form, etc. A computer-readable medium may include any entity or device capable of carrying computer program code, such as a recording medium, a USB disk, a removable hard disk, a magnetic disk, an optical disk, computer memory, a read-only memory, a random access memory, an electrical carrier wave signal, an electrical communication signal, and a software distribution medium.

[0077] The above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the same. The present application has been described in detail with reference to the above examples. However, it should be understood by those skilled in the art that the technical solutions described in each of the above examples can be modified, or some or all of the technical features can be equivalently replaced, and such modifications or replacements will not deviate from the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of each of the embodiments of the present application, and all of them should be included in the protection scope of the present application.

Claims

1. 1. A method for forest disturbance detection based on fully polarized SAR images, comprising: S1: acquiring a full-polarization SAR image within a target area, and performing image segmentation of the full-polarization SAR image into a reference time phase and a monitoring time phase; S2. Performing pre-processing including polarization matrix transformation of the SAR image, refined Lee filtering and false color synthesis on the segmented full polarization SAR image to obtain a false color synthesis SAR image; S3. Input the false color composite SAR image into an optimized ground change detection model to extract ground change patches, and combine them with radar vegetation index (RVI) images to obtain initial forest disturbance patches; S4. Performing a registration check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generating a visualization result; In the step S4, performing an overlap check on the initial forest disturbance patch includes: S411, performing a registration check on the initial forest land disturbance patch and forestry land data to obtain a forestry land disturbance patch; S412. Performing a spatial registration check on the forestry land disturbance patches and slope data to obtain forestry land disturbance patches within a predetermined slope range.

2. The step S2 of performing preprocessing on the divided fully polarized SAR image to obtain a false color composite SAR image specifically includes the following steps: S21. SAR image input: converting the full-polarization SAR images of the reference time phase and the monitoring time phase into a standard data format and performing radiation correction processing to obtain full-polarization single-view composite data X; X={X ｉ :i=1,2,…,M} Here, X ｉ = {x p q ｉ = Ipq ｉ +jQpq ｉ :p, q∈{H, V}} M is the total number of pixels, p is the antenna transmission method, q is the antenna reception method, i is the pixel index, j is the imaginary unit, xpq ｉ is the complex scattering coefficient of the i-th pixel in the pq polarization system, Ipq ｉ xpq ｉ The real part of Qpq ｉ xpq ｉ where H is the horizontal polarization and V is the vertical polarization. S22, polarization matrix transformation of SAR image: constructing a polarization scattering matrix S based on the all-polarization single-view composite data X; S={S ｉ :i=1,2,…,M} defining a target scattering vector set K using the polarization scattering matrix S; K={K ｉ :i=1,2,…,M} converting the set of target scattering vectors K into a polarization coherence matrix T; T={T ｉ :i=1,2,…,M} where * denotes a conjugate operation, < > denotes multi-view processing, S23. Refined Lee filtering: performing refined Lee filtering on the polarization coherent matrix T to obtain a filtered polarization coherent matrix T, including: where N is the number of views, m is the pixel index of the image after multi-view processing, If the resolution in the azimuth direction is changed to 1 / N of the original value and the resolution in the distance direction is not changed, can be obtained, where: denotes the average value of the same boundary matching window within the local window in which the m-th pixel is located, w is the filtering weight value, S24. False color synthesis: performing full polarization false color synthesis on the filtered polarization transformation matrix T1 using the Pauli synthesis method to obtain a false color synthesis SAR image S1 and a false color synthesis SAR image S2. The forest disturbance detection method based on full polarization SAR images according to claim 1,

3. In the step S3, the optimization of the ground surface change detection model is performed as follows: S31, creating a tag image data set based on a ground disturbance interpretation flag; S32, training a ground change detection model using the tag image dataset; S33. Classifying the false-color composite SAR image using a trained ground change detection model to obtain a classification result; S34: performing vector transformation on the classification result, calculating a cross-entropy loss value using a cross-entropy loss function, and performing model evaluation; and S35. if the cross-entropy loss value is less than a cross-entropy loss threshold, output the classification result; otherwise, return to S31 and use the classification result to supplement the tag image data set.

4. In the step S31, the ground disturbance interpretation flag is constructed as follows: S311, acquiring two high-resolution optical remote sensing images, one at an early stage and one at a late stage, which correspond to the time phase of the fully polarized SAR image; S312, inputting the two high-resolution optical remote sensing images from the early and late periods into a pre-training neural network model to extract newly added ground disturbance patches; and S313: combining the false-color composite SAR image with the ground disturbance interpretation flag to construct a ground disturbance interpretation flag.

5. In step S3, the Freeman method is used to calculate the radar vegetation index (RVI); RVI=Fv / (Fv+Fd+Fs) 2. The forest disturbance detection method according to claim 1, wherein Fv, Fd, and Fs respectively represent odd scattering components, even scattering components, and volume scattering components obtained by decomposing the scattering matrix after polarization matrix transformation of the SAR image.

6. In the step S4, generating the visualization result includes: S421, deriving forest disturbance patches within the target area from a vector data layer; S422: Combining with existing attribute information and marking the vector data layer; and S423. generating multiple types of visualization results, including a forest land property classification map, a forest land disturbance heat map, and a forest land disturbance trend graph, based on the marked vector data layer.

7. 1. A forest disturbance detection system based on fully polarized SAR imagery, comprising: a data acquisition module for acquiring a full-polarization SAR image within a target area and performing image segmentation of the full-polarization SAR image into a reference time phase and a monitoring time phase; a data preprocessing module for performing preprocessing on the segmented full-polarization SAR image, including SAR image introduction, SAR image polarization matrix transformation, refined Lee filtering, and false color synthesis, to obtain a false color synthesis SAR image; a feature extraction module for inputting the false-color composite SAR image into an optimized ground change detection model to extract ground change patches, and combining the false-color composite SAR image with a radar vegetation index (RVI) image to obtain initial forest disturbance patches; a registration check module for performing a registration check on the initial forest disturbance patches to obtain forest disturbance patches within the target area and generate a visualization result; performing an overlay check on the initial forest disturbance patch; performing an overlay check on the initial forest disturbance patch and forestry land data to obtain a forestry land disturbance patch; and performing a spatial registration check on the forestry land disturbance patches and slope data to obtain forestry land disturbance patches within a predetermined slope range.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the electronic device realizes the forest disturbance detection method described in any one of claims 1 to 6 when the processor executes the computer program.

9. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the forest disturbance detection method according to any one of claims 1 to 6.

Images