Wide-area landslide identification method and system based on phase gradient stacking and deep learning

Abstract

The present invention provides a wide-area landslide identification method and system based on phase gradient stacking and deep learning. The method comprises: obtaining a set of SAR images covering a study area at different time periods, and using the D-InSAR method to obtain differential interferograms; using the IPGS method to obtain a gradient stack of the differential interferograms, and cropping the gradient stack to obtain a cropped image; labeling the cropped image with landslide samples using auxiliary data and optical images to obtain labeled samples; training an improved YOLOv7 model using the labeled samples to obtain a landslide target recognition model; inputting the image to be identified in the study area into the landslide target recognition model, and outputting the landslide identification results for the study area. The present invention combines the advantages of the proposed improved phase gradient stacking method (rapidly and accurately acquiring deformation gradient signals) with a deep learning model to overcome the shortcomings of existing technologies, thereby providing an efficient and accurate method for automatic identification of potential landslides over a wide area.

Description

Technical Field

[0001] The present invention relates to the field of radar interferometry technology, and in particular to a wide-area landslide identification method and system based on phase gradient stacking and deep learning. Background Art

[0002] Landslides are a widespread and deadly geological hazard that can not only destroy infrastructure such as roads and bridges but also cause significant casualties. With global warming, population growth, and environmental degradation, the frequency and intensity of landslides are increasing significantly. Therefore, research on potential landslide hazards is crucial for reducing casualties, protecting the environment, and promoting sustainable development.

[0003] Synthetic Aperture Radar (SAR) is an Earth observation technology with advantages such as all-weather and all-day operation, wide coverage, and high spatial resolution. Interferometric Synthetic Aperture Radar (InSAR) can capture sub-centimeter-level ground deformation at distances of hundreds of kilometers. After two or three decades of development, time-series InSAR technology has matured and is widely used in landslide identification and monitoring, providing crucial support for landslide hazard monitoring, mechanism analysis, and physical inversion research. However, the use of time-series InSAR for wide-area landslide identification has limitations, particularly in complex mountainous areas. Time-series results are often affected by severe atmospheric delay and potential phase unwrapping errors. Furthermore, the time-series results require unwrapping and subsequent time-series processing, which in turn requires significant computational effort. Consequently, atmospheric delay, unwrapping errors, and the sheer volume of computations required significantly reduce the ability of InSAR to detect landslides in high mountainous areas. With the advent of the era of massive SAR data, the computational burden of InSAR is exponentially increasing, creating an urgent need for fast and accurate InSAR landslide identification technology. Therefore, how to effectively process large-scale InSAR data and quickly and automatically interpret the results is an urgent problem that needs to be solved. Summary of the Invention

[0004] The present invention provides a wide-area landslide identification method and system based on phase gradient stacking and deep learning, which are used to solve the defects in the landslide disaster identification methods in the prior art.

[0005] In a first aspect, the present invention provides a wide-area landslide identification method based on phase gradient stacking and deep learning, comprising:

[0006] A set of SAR images covering the study area at different time periods was obtained, and differential interferograms were obtained using the D-InSAR method;

[0007] Obtaining a gradient stacking image of the differential interferogram using an IPGS method, and cropping the gradient stacking image to obtain a cropped image;

[0008] Using auxiliary data and optical images, labeling the cropped image with landslide samples to obtain labeled samples;

[0009] The labeled samples are used to train the improved YOLOv7 model to obtain a landslide target recognition model;

[0010] The image to be identified in the study area is input into the landslide target identification model, and the landslide identification result of the study area is output.

[0011] According to a wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention, an IPGS method is used to obtain a gradient stacking image of the differential interferogram, and the gradient stacking image is cropped to obtain a cropped image, comprising:

[0012] Calculating the phase gradient in any direction of the differential interferogram based on a preset step size;

[0013] The phase gradients of each differential interferogram corresponding to any direction are stacked from the time dimension and median filtered;

[0014] Merge the phase stacking images in each direction and perform gradient normalization on the merged results;

[0015] Converting the gradient normalization result to a geographic coordinate system to obtain the gradient stacking map;

[0016] The gradient stacked image is cropped into an image of a preset size to obtain the cropped image.

[0017] According to a wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention, landslide samples are annotated on the cropped image using auxiliary data and optical images to obtain annotated samples, including:

[0018] Cropping the auxiliary data into a picture of a preset size to obtain cropped auxiliary data, and obtaining a slope map;

[0019] The cropped image is annotated using a preset annotation tool in combination with the optical image, the slope map and the cropped auxiliary data to obtain the annotated sample.

[0020] According to a wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention, the improved YOLOv7 model is obtained by the following steps:

[0021] Determining that the improved YOLOv7 model includes an input end, a backbone network, a head module, and a prediction layer;

[0022] The backbone network includes a CBS module, an ELAN module and an MP module;

[0023] The head module includes an SPPCSPC module, a CBS module, an ELAN module and an MP module, forming an FPN network structure;

[0024] The prediction module includes three detection layers, which enable the network to detect objects of different sizes from three different receptive fields.

[0025] The CBAM attention mechanism is introduced into the backbone network to capture the correlation between features by adaptively learning channels and spatial attention weights.

[0026] According to a wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention, the improved YOLOv7 model is trained using the labeled samples to obtain a landslide target recognition model, including:

[0027] Use the weights trained with the preset image dataset as the network's pre-trained weights to determine the batch size and epoch.

[0028] The improved YOLOv7 model is trained using the labeled samples, and the model performance is evaluated using target detection performance indicators. The landslide target recognition model is obtained after training convergence.

[0029] According to a wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention, after the improved YOLOv7 model is trained using the labeled samples to obtain a landslide target identification model, the method further includes:

[0030] If the study area is determined to be a snow mountain area, the snow cover normalized difference index is used to obtain the snow cover area of ​​the study area, and the snow mountain area is masked out from the gradient stacking map.

[0031] In a second aspect, the present invention further provides a wide-area landslide identification system based on phase gradient stacking and deep learning, comprising:

[0032] The acquisition module is used to obtain a set of SAR images covering the study area at different time periods and obtain differential interferograms using the D-InSAR method;

[0033] a stacking module, configured to obtain a gradient stacking image of the differential interferogram using an IPGS method, and to crop the gradient stacking image to obtain a cropped image;

[0034] a labeling module, configured to label the cropped image with landslide samples using auxiliary data and optical images to obtain labeled samples;

[0035] A training module, configured to train the improved YOLOv7 model using the labeled samples to obtain a landslide target recognition model;

[0036] The identification module is used to input the image to be identified in the study area into the landslide target identification model and output the landslide identification result of the study area.

[0037] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the wide-area landslide identification method based on phase gradient stacking and deep learning as described above is implemented.

[0038] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the wide-area landslide identification methods based on phase gradient stacking and deep learning as described above.

[0039] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described wide-area landslide identification methods based on phase gradient stacking and deep learning.

[0040] The present invention provides a wide-area landslide identification method and system based on phase gradient stacking and deep learning. By using the IPGS method to obtain gradient stacking results for the study area, the system significantly reduces the computational effort compared to traditional InSAR methods. Combining these results with deep learning allows for the timely and automatic identification of potential landslides. The obtained gradient stacking results and corresponding auxiliary data (slope) are used as input data. Furthermore, the present invention adds input data channels to the YOLOv7 model to accommodate multiple data inputs. Furthermore, the CBAM attention mechanism is added to the network model to improve network performance.

[0041] The present invention uses phase gradient stacking results and auxiliary data as input to the YOLOv7 model, which reduces the workload on the one hand and improves the accuracy of potential landslide identification on the other. The introduction of the attention mechanism enhances the recognition ability of the model, thereby improving the performance of the network model. The IPGS method not only effectively suppresses the influence of atmospheric delay noise, but also can quickly obtain gradient stacking results in a large-scale study area, thereby enabling a timely survey of geological disasters. The obtained gradient stacking results and auxiliary data (slope) are used as input data. An input data channel is added to the YOLOv7 model to accommodate multiple data inputs; and the CBAM attention mechanism is added to the network model to improve network performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 This is one of the flow charts of the wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention;

[0044] Figure 2 This is the second flow chart of the wide-area landslide identification method based on phase gradient stacking and deep learning provided by the present invention;

[0045] Figure 3 It is a YOLOv7 network model structure diagram provided by the present invention;

[0046] Figure 4 This is a schematic diagram of the CBAM attention mechanism structure provided by the present invention;

[0047] Figure 5 This is a potential landslide identification result map of a certain province provided by the present invention;

[0048] Figure 6 Schematic diagram of the structure of the wide-area landslide identification system based on phase gradient stacking and deep learning provided by the present invention;

[0049] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0050] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0051] Figure 1 This is one of the flow charts of the wide-area landslide identification method based on phase gradient stacking and deep learning provided by an embodiment of the present invention, such as Figure 1 As shown, including:

[0052] Step 100: Obtain a set of SAR images covering the study area at different time periods, and obtain differential interferograms using the D-InSAR method;

[0053] Step 200: using the IPGS method to obtain a gradient stacking image of the differential interferogram, and cropping the gradient stacking image to obtain a cropped image;

[0054] Step 300: labeling the landslide samples on the cropped image using auxiliary data and optical images to obtain labeled samples;

[0055] Step 400: using the labeled samples to train the improved YOLOv7 model to obtain a landslide target recognition model;

[0056] Step 500: input the image to be identified in the study area into the landslide target identification model, and output the landslide identification result of the study area.

[0057] Specifically, the embodiment of the present invention adopts a wide-area potential landslide automatic identification method based on phase gradient stacking and deep learning, which supports wide-area landslide automatic identification, such as Figure 2 As shown, the implementation process includes:

[0058] SAR images covering the study area at different time periods were acquired and processed using the conventional Differential Interferometric Synthetic Aperture Radar (D-InSAR) method to obtain differential interferograms. An improved phase gradient stacking (IPGS) method was used to obtain gradient stacking maps of the study area and crop the images. With the assistance of auxiliary data and optical images, the cropped images were annotated with landslide samples. For the target detection model YOLOv7, a multi-channel input window was added and an attention mechanism was introduced into the backbone network to improve the performance of the network model in landslide recognition. The annotated images and the corresponding auxiliary images were input into the target recognition model for training and testing. The trained model was used to carry out wide-area potential landslide identification.

[0059] Based on the above embodiment, step 200 includes:

[0060] Calculating the phase gradient in any direction of the differential interferogram based on a preset step size;

[0061] The phase gradients of each differential interferogram corresponding to any direction are stacked from the time dimension and median filtered;

[0062] Merge the phase stacking images in each direction and perform gradient normalization on the merged results;

[0063] Converting the gradient normalization result to a geographic coordinate system to obtain the gradient stacking map;

[0064] The gradient stacked image is cropped into an image of a preset size to obtain the cropped image.

[0065] Specifically, an embodiment of the present invention calculates the phase gradient of an interference pattern in a certain direction with a preset step size s; stacks the phase gradient of each interference pattern corresponding to a certain direction in the time dimension and performs median filtering; merges and normalizes the phase stacking results in each direction; and converts the phase gradient stacking results to a geographic coordinate system to obtain a gradient stacking map.

[0066] In this embodiment, a certain province is taken as the study area, and the data source is the European Space Agency's Sentinel-1 data covering the province. A total of about 116 SAR satellite image data are obtained from January 2018 to February 2022.

[0067] Furthermore, the IPGS method was used to obtain a gradient stacking map of the study area and crop the image. The phase gradient of a specific direction in the interferogram was calculated with a preset step size s. The phase gradient corresponding to a specific direction in each interferogram was stacked in the time dimension and median filtered. The phase stacking results for each direction were merged and gradient normalized. The result was then converted to a geographic coordinate system. Accordingly, the slope corresponding to the gradient result was calculated using the DEM data. Finally, the gradient stacking result was cropped into a 640×640 image.

[0068] Based on the above embodiment, step 300 includes:

[0069] Cropping the auxiliary data into a picture of a preset size to obtain cropped auxiliary data, and obtaining a slope map;

[0070] The cropped image is annotated using a preset annotation tool in combination with the optical image, the slope map and the cropped auxiliary data to obtain the annotated sample.

[0071] Specifically, the embodiment of the present invention annotates landslide samples on a cropped portion of the image with the aid of auxiliary data and optical images.

[0072] Because landslide occurrence is significantly correlated with slope, the corresponding optical image and slope map were considered when annotating potential landslides on the cropped gradient stack image. Auxiliary data was also cropped to 640×640 pixels, and potential landslides were annotated using the labelimg annotation tool.

[0073] Based on the above embodiments, the improved YOLOv7 model constructed in the embodiments of the present invention adds a multi-channel input window to the target detection model YOLOv7 and introduces an attention mechanism in the backbone network to improve the performance of the network model in landslide recognition.

[0074] like Figure 3As shown in the figure, the YOLOv7 model consists of four main components: the input, the backbone network, the head module, and the detection layer. The backbone network primarily consists of the CBS module, the ELAN module, and the MP module. The head module primarily comprises the SPPCSPC, CBS, ELAN, and MP modules, forming the FPN network structure. The prediction layer includes three detection layers, allowing the network to detect objects of different sizes from three different receptive fields. A multi-channel input window is added to incorporate auxiliary data (slope data) to improve the accuracy of landslide identification. Furthermore, the CBAM attention mechanism is introduced on top of the backbone network. CBAM adaptively learns channel and spatial attention weights to capture correlations between features, thereby improving the performance of image recognition tasks.

[0075] Since the occurrence of landslides has a certain correlation with slope, the introduction of slope data during network model training makes the network training have a certain direction, which significantly improves the accuracy of network landslide identification and also reduces the model's inference time.

[0076] The ELAN module achieves deep learning and convergence of the network by controlling the shortest and longest gradient paths. The MP structure combines two downsampling techniques, pooling and convolution, enabling the network to select the best downsampling method between the two. The role of SPPCSPC is to use the maximum pooling of four different scales for processing to achieve the fusion of information of different feature scales. Its main function is to perform multi-scale feature fusion after the backbone network feature extraction and pass these features to the prediction layer. The end of the head module consists of REP and CBM modules. REP is used to adjust the number of channels of different scale features of the backbone network, extract features, and then use CBM to predict bounding boxes and target categories. The prediction layer includes three detection layers, which enable the network to detect targets of different sizes from three receptive fields respectively.

[0077] For the application of the single-stage target detection model YOLOv7 in potential landslide identification, the present invention introduces the CBAM attention mechanism based on the backbone network, which enhances the network's ability to effectively extract deep features and effectively improves the network's ability to identify small landslides.

[0078] CBAM is an attention mechanism that combines channel attention and spatial attention modules. By adaptively assigning weights to different channels and spatial positions of the feature map, the convolutional neural network's attention to important features is enhanced, thereby improving the performance of the model. The CBAM module has a simple structure but powerful functions and is divided into two sub-modules: the channel attention module and the spatial attention module. The channel attention module extracts features through global average pooling and maximum pooling, and processes them through a shared multi-layer perceptron to generate a channel attention map. The spatial attention module processes the channel-weighted feature map and generates a spatial attention map through convolution operations. The main feature of CBAM is that it can pay attention to channel and spatial information at the same time. This combination of dual attention mechanisms enables the network to extract and utilize key information more accurately when processing complex visual tasks. The module structure diagram is shown below. Figure 4 shown.

[0079] Based on the above embodiment, step 400 includes:

[0080] Use the weights trained with the preset image dataset as the network's pre-trained weights to determine the batch size and epoch.

[0081] The improved YOLOv7 model is trained using the labeled samples, and the model performance is evaluated using target detection performance indicators. The landslide target recognition model is obtained after training convergence.

[0082] Specifically, in this embodiment of the present invention, weights trained with COCO data are used as pre-trained weights for the network; batch_size is set to 8; and epoch is set to 300. This model is evaluated using the commonly used performance indicators in object detection algorithms: recall, precision, and average precision (AP).

[0083] In addition, the experimental environment of the present invention is: system: Windows 10; processor Intel(R) Core(TM) i7-10700CPU@2.90GHz; memory: 64G; graphics card: NVIDIA TITAN RTX; deep learning framework is Pytorch version 1.12; compilation language is Python version 3.8; CUDA version is 11.2; Cudnn version is 8.3.0.

[0084] Based on the above embodiment, after step 400, the following steps are further included:

[0085] If the study area is determined to be a snow mountain area, the snow cover normalized difference index is used to obtain the snow cover area of ​​the study area, and the snow mountain area is masked out from the gradient stacking map.

[0086] Specifically, after completing the training of the model, the embodiment of the present invention inputs the phase gradient stacking map and auxiliary data into the trained model to carry out automatic identification of potential landslides in a wide area. Since the deformation caused by the melting of snow in the snow-capped mountain area is similar to the landslide deformation, if the study area is covered with snow-capped mountains, optical images of the study area will be collected, and the Normalized Difference Snow Index (NDSI) will be used to obtain the snow-covered area of ​​the study area and mask out the snow-capped mountain area from the IPGS results before automatic landslide identification is carried out to finally obtain the landslide identification results of the study area. Figure 5 This is the spatial distribution of potential landslides identified in a certain province by the present invention.

[0087] Finally, using the landslide target recognition model obtained by the present invention, the user only needs to provide a gradient stacking map and corresponding auxiliary images, and the system can automatically identify potential landslides based on the trained model.

[0088] The wide-area landslide identification system based on phase gradient stacking and deep learning provided by the present invention is described below. The wide-area landslide identification system based on phase gradient stacking and deep learning described below and the wide-area landslide identification method based on phase gradient stacking and deep learning described above can be referenced to each other.

[0089] Figure 6 Schematic diagram of the structure of the wide-area landslide identification system based on phase gradient stacking and deep learning provided by an embodiment of the present invention. Figure 6 As shown, it includes: an acquisition module 61, a stacking module 62, a labeling module 63, a training module 64 and a recognition module 65, wherein:

[0090] The acquisition module 61 is used to obtain a set of SAR images covering different time periods in the study area, and obtain a differential interference map using the D-InSAR method; the stacking module 62 is used to obtain a gradient stacking map of the differential interference map using the IPGS method, and crop the gradient stacking map to obtain a cropped image; the labeling module 63 is used to label the cropped image with landslide samples using auxiliary data and optical images to obtain labeled samples; the training module 64 is used to train the improved YOLOv7 model using the labeled samples to obtain a landslide target recognition model; the recognition module 65 is used to input the image to be identified in the study area into the landslide target recognition model, and output the landslide recognition result of the study area.

[0091] Figure 7 An example of a physical structure diagram of an electronic device is shown below. Figure 7As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 may call logic instructions in the memory 730 to execute a wide-area landslide identification method based on phase gradient stacking and deep learning. The method includes: obtaining a set of SAR images covering different time periods of a study area, and obtaining a differential interferogram using the D-InSAR method; obtaining a gradient stacking image of the differential interferogram using the IPGS method, and cropping the gradient stacking image to obtain a cropped image; labeling the cropped image with landslide samples using auxiliary data and optical images to obtain labeled samples; using the labeled samples to train an improved YOLOv7 model to obtain a landslide target recognition model; inputting the image to be identified in the study area into the landslide target recognition model, and outputting a landslide identification result for the study area.

[0092] Furthermore, the logic instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0093] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the wide-area landslide identification method based on phase gradient stacking and deep learning provided by the above methods. The method includes: obtaining a set of SAR images covering different time periods of the study area, and obtaining a differential interference map using the D-InSAR method; obtaining a gradient stacking map of the differential interference map using the IPGS method, and cropping the gradient stacking map to obtain a cropped image; using auxiliary data and optical images, annotating the cropped image with landslide samples to obtain annotated samples; using the annotated samples to train an improved YOLOv7 model to obtain a landslide target recognition model; inputting the image to be identified in the study area into the landslide target recognition model, and outputting the landslide identification result of the study area.

[0094] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the wide-area landslide identification method based on phase gradient stacking and deep learning provided by the above-mentioned methods, the method comprising: obtaining a set of SAR images covering different time periods of the study area, and obtaining a differential interference map using the D-InSAR method; obtaining a gradient stacking map of the differential interference map using the IPGS method, and cropping the gradient stacking map to obtain a cropped image; labeling the cropped image with landslide samples using auxiliary data and optical images to obtain labeled samples; using the labeled samples to train an improved YOLOv7 model to obtain a landslide target recognition model; inputting the image to be identified in the study area into the landslide target recognition model, and outputting the landslide identification result of the study area.

[0095] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0096] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A wide-area landslide identification method based on phase gradient stacking and deep learning, characterized in that: include: A set of synthetic aperture radar (SAR) images covering the study area at different times was obtained, and differential interferograms were obtained using the differential synthetic aperture radar interferometry (D-InSAR) method. An improved phase gradient stacking IPGS method is used to obtain a gradient stacking image of the differential interferogram, and the gradient stacking image is cropped to obtain a cropped image; Using the slope map and the optical image, marking the landslide samples on the cropped image to obtain marked samples; The slope map and the labeled samples are used to train an improved YOLOv7 model to obtain a landslide target recognition model; Inputting the image to be identified in the study area and the slope map of the study area into the landslide target identification model, and outputting the landslide identification result of the study area; The IPGS method is used to obtain a gradient stacking image of the differential interferogram, and the gradient stacking image is cropped to obtain a cropped image, including: Calculating the phase gradient in any direction of the differential interferogram based on a preset step size; The phase gradients of each differential interferogram corresponding to any direction are stacked from the time dimension and median filtered; Merge the phase stacking images in each direction and perform gradient normalization on the merged results; Converting the gradient normalization result to a geographic coordinate system to obtain the gradient stacking map; Cropping the gradient stacked image into an image of a preset size to obtain the cropped image; The landslide sample annotation is performed on the cropped image using the slope map and the optical image to obtain an annotation sample, including: Cropping the slope map into a preset size image to obtain a cropped slope map; The cropped image is annotated using a preset annotation tool in combination with the optical image and the cropped slope map to obtain the annotated sample.

2. The wide-area landslide identification method based on phase gradient stacking and deep learning according to claim 1 is characterized in that: The improved YOLOv7 model is obtained by the following steps: Determining that the improved YOLOv7 model includes an input end, a backbone network, a head module, and a prediction layer; The backbone network includes a CBS module, an ELAN module and an MP module; The head module includes an SPPCSPC module, a CBS module, an ELAN module and an MP module, forming an FPN network structure; The prediction module includes three detection layers, which enable the network to detect objects of different sizes from three different receptive fields. The CBAM attention mechanism is introduced into the backbone network to capture the correlation between features by adaptively learning channels and spatial attention weights.

3. The wide-area landslide identification method based on phase gradient stacking and deep learning according to claim 1 is characterized in that: The slope map and the labeled samples are used to train the improved YOLOv7 model to obtain a landslide target recognition model, including: Use the weights trained with the preset image dataset as the network's pre-trained weights to determine the batch size and epoch. The improved YOLOv7 model is trained using the slope map and the labeled samples, and the model performance is evaluated using target detection performance indicators. The landslide target recognition model is obtained after training convergence.

4. The wide-area landslide identification method based on phase gradient stacking and deep learning according to claim 1 is characterized in that: After the improved YOLOv7 model is trained using the slope map and the labeled samples to obtain a landslide target recognition model, the method further includes: If the study area is determined to be a snow mountain area, the snow cover normalized difference index is used to obtain the snow cover area of ​​the study area, and the snow mountain area is masked out from the gradient stacking map.

5. A wide-area landslide identification system based on phase gradient stacking and deep learning, based on the wide-area landslide identification method based on phase gradient stacking and deep learning according to any one of claims 1 to 4, characterized in that: include: The acquisition module is used to obtain a set of SAR images covering the study area at different time periods and obtain differential interferograms using the D-InSAR method; a stacking module, configured to obtain a gradient stacking image of the differential interferogram using an IPGS method, and to crop the gradient stacking image to obtain a cropped image; a labeling module, configured to label the cropped image with landslide samples using the slope map and the optical image to obtain labeled samples; A training module, configured to train an improved YOLOv7 model using the slope map and the labeled samples to obtain a landslide target recognition model; The identification module is used to input the image to be identified in the study area and the slope map of the study area into the landslide target identification model, and output the landslide identification result of the study area.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the wide-area landslide identification method based on phase gradient stacking and deep learning as described in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the wide-area landslide identification method based on phase gradient stacking and deep learning as claimed in any one of claims 1 to 4 is implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the wide-area landslide identification method based on phase gradient stacking and deep learning as claimed in any one of claims 1 to 4 is implemented.

Images