A landslide area detection method and system based on radar remote sensing data

Abstract

The present application provides a method and system for detecting landslide areas based on radar remote sensing data. The method extracts the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area at each time phase from the radar remote sensing data. The scattering characteristic matrix constructed based on the backscatter coefficient at each time phase is combined with the spatiotemporal variation characteristics of the coherence coefficient to determine the surface deformation sensitivity of each pixel point in the target scanning area. The deformation-sensitive area and stability partition are divided according to the singular value decomposition results of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point. The hidden landslide points in the target scanning area are identified from the stability partition based on the scattering entropy of the pixel points in the deformation-sensitive area. Then, a landslide risk probability map is generated by combining all the hidden landslide points with the deformation-sensitive area. The technical solution provided by the present application can effectively identify hidden landslide points in the scanning area when weak deformation signals are submerged.

Description

Technical Field

[0001] The present application relates to the field of radio detection technology, and more specifically, to a landslide area detection method and system based on radar remote sensing data. Background Art

[0002] With the rapid development of modern science and technology, radio plays a key role in many fields, among which radar radio detection technology is particularly important. Radar detects information such as the position, speed, and shape of target objects by emitting radio waves and receiving their reflected signals. However, with the increasing complexity of the electromagnetic environment and the increase in interference and clutter, higher requirements are placed on the accuracy and reliability of radar radio detection, prompting continuous innovation in related technologies to improve their performance and adaptability.

[0003] In existing radio detection, radio detection is mainly based on the propagation characteristics of electromagnetic waves. Radar radio detection is a typical example. Radar radio detection transmits electromagnetic waves of a specific frequency. When these electromagnetic waves encounter the target object, they will be reflected. The radar receives the reflected electromagnetic wave signal and determines the target's position, speed and distance information based on the signal's propagation time and frequency changes, thereby realizing target detection. However, in the detection of landslide areas based on radar remote sensing data, the surface deformation in the early stage of the landslide is weak and has nonlinear characteristics. The single deformation parameter relied on by traditional methods is easily interfered by environmental noise (such as vegetation cover and humidity changes), resulting in the weak deformation signal being submerged, making it difficult to distinguish between real deformation and pseudo-change areas, resulting in the inability to effectively identify hidden landslide points in the scanned area, thereby reducing the effectiveness of landslide area detection. Therefore, how to effectively identify hidden landslide points in the scanned area when the weak deformation signal is submerged, and thus improve the effectiveness of landslide area detection, has become a difficult problem faced by the industry. Summary of the Invention

[0004] The present application provides a landslide area detection method and system based on radar remote sensing data, which can effectively identify hidden landslide points in the scanning area when weak deformation signals are submerged, thereby improving the effectiveness of landslide area detection.

[0005] In a first aspect, the present application provides a method for detecting landslide areas based on radar remote sensing data, comprising the following steps:

[0006] Acquire radar remote sensing data of the target scanning area in multiple time phases;

[0007] Extract the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area in each phase from the corresponding radar remote sensing data;

[0008] Constructing a scattering characteristic matrix of the surface medium in the target scanning area to radar waves based on the backscatter coefficients in each time phase, and determining the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient;

[0009] Performing singular value decomposition on the scattering characteristic matrix, and dividing the target scanning area into deformation sensitive areas and stability zones according to the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point;

[0010] The scattering entropy of the pixel points in the deformation sensitive area is determined, and the hidden landslide points in the target scanning area are identified from the stability partition according to the scattering entropy, and then a landslide risk probability map is generated by combining all the hidden landslide points with the deformation sensitive area.

[0011] In some embodiments, extracting the backscatter coefficient and the coherence coefficient when the detection radar detects the target scanning area at each time phase from the corresponding radar remote sensing data specifically includes:

[0012] Preprocess the radar remote sensing data at each time phase to generate single-view complex radar images of the target scanning area at each time phase;

[0013] A time phase is selected as a selected time phase, and the backscatter coefficient of the detection radar detecting the target scanning area in the selected time phase is extracted from the single-look complex radar image corresponding to the selected time phase according to the radiation calibration function;

[0014] Registering the single-view complex radar image corresponding to the selected time with the single-view complex radar image corresponding to the adjacent time to generate a single-view complex radar image pair corresponding to the selected time;

[0015] Determining, based on the single-view complex radar image pair, a coherence coefficient when the detection radar detects the target scanning area at a selected time phase;

[0016] Continue to determine the backscatter coefficient and coherence coefficient when the detection radar detects the target scanning area in the remaining time phase.

[0017] In some embodiments, constructing a scattering characteristic matrix of the surface medium to the radar wave in the target scanning area based on the backscatter coefficients in each time phase specifically includes:

[0018] Obtain the backscatter coefficient at each time phase. Different pixel points in the target scanning area correspond to a backscatter coefficient.

[0019] Arrange the backscatter coefficients at each time phase in time series to generate a backscatter coefficient vector at each pixel point in the target scanning area;

[0020] The backscattering coefficient vectors at all pixel points are combined into a two-dimensional matrix according to the spatial dimension to obtain the scattering characteristic matrix of the surface medium to the radar wave in the target scanning area.

[0021] In some embodiments, determining the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient specifically includes:

[0022] Determining the scattering fluctuation intensity corresponding to each pixel point based on the backscattering coefficient vector corresponding to each pixel point in the scattering characteristic matrix;

[0023] Obtain the coherence coefficient sequence of each pixel in all time intervals;

[0024] Determining the spatiotemporal variation characteristics corresponding to each pixel point according to the coherence coefficient sequence corresponding to each pixel point;

[0025] For each pixel point, the scattering fluctuation intensity corresponding to the pixel point is normalized and weighted fused with the spatiotemporal variation characteristics to obtain the surface deformation sensitivity of each pixel point in the target scanning area.

[0026] In some embodiments, dividing the target scanning area into deformation-sensitive areas and stability zones based on the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point specifically includes:

[0027] extracting principal singular vectors from a singular value decomposition result of the scattering characteristic matrix;

[0028] Determine the principal component projection value of each pixel point according to the principal singular vector combined with the backscattering coefficient vector of each pixel point in the scattering characteristic matrix;

[0029] The characteristic response map of the surface deformation in the target scanning area is determined by the principal component projection value corresponding to each pixel point and the sensitivity of the surface deformation;

[0030] The target scanning area is divided into deformation sensitive areas and stability zones based on the characteristic response map.

[0031] In some embodiments, determining the scattering entropy of the pixel points in the deformation sensitive area specifically includes:

[0032] Extract the backscattering coefficient sequence of each pixel in the deformation sensitive area;

[0033] A pixel point is selected as a selected pixel point, and the scattering entropy of the selected pixel point is determined according to a backscattering coefficient sequence of the selected pixel point;

[0034] Continue to determine the scattering entropy of the remaining pixel points in the deformation sensitive area.

[0035] In some embodiments, radar remote sensing data of the target scanning area in multiple time phases is acquired by a detection radar.

[0036] In a second aspect, the present application provides a landslide area detection system based on radar remote sensing data, which is used to perform a landslide area detection method based on radar remote sensing data, including:

[0037] An acquisition module is used to acquire radar remote sensing data of the target scanning area in multiple time phases;

[0038] A processing module is used to extract the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area in each time phase from the corresponding radar remote sensing data;

[0039] The processing module is further configured to construct a scattering characteristic matrix of the surface medium in the target scanning area to the radar wave based on the backscattering coefficients in each time phase, and determine the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient;

[0040] The processing module is further configured to perform singular value decomposition on the scattering characteristic matrix, and divide the target scanning area into deformation-sensitive areas and stability zones based on the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point;

[0041] An execution module is used to determine the scattering entropy of the pixel points in the deformation sensitive area, identify the hidden landslide points in the target scanning area from the stability partition according to the scattering entropy, and then generate a landslide risk probability map by combining all the hidden landslide points with the deformation sensitive area.

[0042] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned landslide area detection method based on radar remote sensing data.

[0043] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned landslide area detection method based on radar remote sensing data.

[0044] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0045] The landslide area detection method and system based on radar remote sensing data provided in the present application are as follows: first, radar remote sensing data of a target scanning area in multiple time phases are acquired; second, backscattering coefficients and coherence coefficients of the detection radar when detecting the target scanning area in each time phase are extracted from the corresponding radar remote sensing data; further, a scattering characteristic matrix of the surface medium in the target scanning area to radar waves is constructed based on the backscattering coefficients in each time phase, and the surface deformation sensitivity of each pixel point in the target scanning area is determined according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient; then, singular value decomposition is performed on the scattering characteristic matrix, and deformation sensitive areas and stability partitions in the target scanning area are divided according to the singular value decomposition results of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point; finally, the scattering entropy of the pixel points in the deformation sensitive area is determined, and the hidden landslide points in the target scanning area are identified from the stability partitions according to the scattering entropy, and then a landslide risk probability map is generated by combining all the hidden landslide points with the deformation sensitive area.

[0046] It can be seen that the present application can effectively identify hidden landslide points in the scanning area when weak deformation signals are submerged, thereby improving the effectiveness of landslide area detection; first, the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area in each phase are extracted from the corresponding radar remote sensing data, which can effectively identify signs of surface deformation or landslide and reflect the stability and degree of change of the surface scattering characteristics; secondly, based on the backscatter coefficient in each phase, a scattering characteristic matrix of the surface medium in the target scanning area to radar waves is constructed, and the surface deformation sensitivity of each pixel point in the target scanning area is determined according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient, so as to measure the degree of sensitivity of each pixel position in the target scanning area to surface deformation response, and then effectively identify those surface areas that show obvious abnormalities or instability in radar wave reflection characteristics, thereby avoiding interference from environmental noise. the problem of deformation signal being submerged; further, the scattering characteristic matrix is ​​subjected to singular value decomposition, and the deformation sensitive area and stability partition in the target scanning area are divided according to the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point, so as to identify the area with potential geological disaster risk in advance, provide data support for subsequent landslide risk assessment, and thus realize the advancement and precision of disaster prevention and control work; then, according to the scattering entropy of the pixel points in the deformation sensitive area, the hidden landslide points in the target scanning area are identified from the stability partition to obtain the hidden landslide risk area, so as to avoid the landslide risk area with weak surface disturbance from being missed; finally, a landslide risk probability map is generated by combining all the hidden landslide points with the deformation sensitive area; in summary, the technical solution provided by the present application can effectively identify the hidden landslide points in the scanning area when the weak deformation signal is submerged, thereby improving the effectiveness of landslide area detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is an exemplary flow chart of a landslide area detection method based on radar remote sensing data according to some embodiments of the present application;

[0048] Figure 2 is an exemplary flow chart of determining a scattering characteristic matrix according to some embodiments of the present application;

[0049] Figure 3 is an exemplary flow chart for determining deformation sensitive areas and stability zones according to some embodiments of the present application;

[0050] Figure 4 is a schematic structural diagram of a landslide area detection system based on radar remote sensing data according to some embodiments of the present application;

[0051] Figure 5 It is a structural diagram of a computer device for implementing a landslide area detection method based on radar remote sensing data according to some embodiments of the present application. DETAILED DESCRIPTION

[0052] In order to better understand the technical solution of the present application, the technical solution of the present application will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0053] refer to Figure 1 , which is an exemplary flow chart of a landslide area detection method based on radar remote sensing data according to some embodiments of the present application. The landslide area detection method 100 based on radar remote sensing data mainly includes the following steps:

[0054] In step S101 , radar remote sensing data of a target scanning area in multiple time phases is acquired.

[0055] In a specific implementation, radar remote sensing data of the target scanning area in multiple time phases is obtained by a detection radar. The multiple time phases refer to multiple time points. The radar remote sensing data represents the observation data obtained after the detection radar scans the target scanning area. The radar remote sensing data is radar remote sensing image data, wherein the detection radar is a synthetic aperture radar.

[0056] It should be noted that the target scanning area in this application represents a landslide area where landslide risks may occur.

[0057] In step S102, the backscatter coefficient and the coherence coefficient when the detection radar detects the target scanning area in each time phase are extracted from the corresponding radar remote sensing data.

[0058] In some embodiments, extracting the backscatter coefficient and coherence coefficient of the detection radar detecting the target scanning area at each time phase from the corresponding radar remote sensing data can be achieved by using the following steps:

[0059] Preprocess the radar remote sensing data at each time phase to generate single-view complex radar images of the target scanning area at each time phase;

[0060] A time phase is selected as a selected time phase, and the backscatter coefficient of the detection radar detecting the target scanning area in the selected time phase is extracted from the single-look complex radar image corresponding to the selected time phase according to the radiation calibration function;

[0061] Registering the single-view complex radar image corresponding to the selected time with the single-view complex radar image corresponding to the adjacent time to generate a single-view complex radar image pair corresponding to the selected time;

[0062] Determining, based on the single-view complex radar image pair, a coherence coefficient when the detection radar detects the target scanning area at a selected time phase;

[0063] Continue to determine the backscatter coefficient and coherence coefficient when the detection radar detects the target scanning area in the remaining time phase.

[0064] It should be noted that the backscatter coefficient in this application is used to measure the backscattering ability of the surface to radar waves. The backscatter coefficients obtained at different times in the same area will change due to factors such as soil moisture, vegetation status, and structural changes. Therefore, the determination of the backscatter coefficient can be used to identify surface deformation or signs of landslides; the coherence coefficient in this application is an indicator that measures the phase consistency of two radar images obtained at different times at the same pixel position. The coherence coefficient reflects the stability and degree of change of the surface scattering characteristics, and is an important characteristic parameter for identifying dynamic changes such as surface deformation and landslide activities.

[0065] In the specific implementation, first, the existing radiation correction technology is used to preprocess the radar remote sensing data in each phase, and the radar remote sensing data is a radar remote sensing image, and the preprocessed radar remote sensing image is used as the single-view complex radar image of the target scanning area in the corresponding phase; secondly, a phase is selected as the selected phase, and the grayscale value at each pixel point in the single-view complex radar image corresponding to the selected phase is input as an input variable into the radiation calibration function, and the output result of the radiation calibration function is used as the backscattering coefficient at the pixel point, and then the backscattering coefficient when the detection radar detects the target scanning area in the selected phase is extracted, and the radiation calibration function refers to a mathematical relationship for converting the original digital value of the radar remote sensing image into a radar backscattering coefficient with physical meaning; further, the single-view complex radar image corresponding to the selected phase is used as the input variable. The single-look complex radar images corresponding to the adjacent time are aligned to generate a single-look complex radar image pair corresponding to the selected time; then, the coherence coefficient of the detection radar when detecting the target scanning area at the selected time phase is determined based on the single-look complex radar image pair, that is: for the pixel point at the same position in the single-look complex radar image pair, with the pixel point as the center and a preset sliding window as the range, the mutual correlation coefficient of the two single-look complex radar images in the sliding window is used as the coherence coefficient of the pixel point, and then the coherence coefficient of the detection radar when detecting the target scanning area at the selected time phase is obtained; finally, the backscattering coefficient and coherence coefficient of the detection radar when detecting the target scanning area at the remaining time phases are continued to be determined according to the determination method of "determining the backscattering coefficient and coherence coefficient when the detection radar detects the target scanning area at the selected time phase", which will not be repeated here.

[0066] It should be noted that, in this embodiment, a single-look complex radar image refers to an image that retains the complex information (amplitude and phase) of each pixel in the radar echo data after single-view processing. It is the most basic and sophisticated image format in the synthetic aperture radar processing flow. In this embodiment, a single-look complex radar image pair represents a combination of two adjacent single-look complex radar images.

[0067] In step S103, a scattering characteristic matrix of the surface medium in the target scanning area to radar waves is constructed based on the backscattering coefficients in each time phase, and the surface deformation sensitivity of each pixel point in the target scanning area is determined according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient.

[0068] In some embodiments, reference Figure 2 As shown in FIG. 1 , this figure is an exemplary flow chart for determining a scattering characteristic matrix according to some embodiments of the present application. In this embodiment, constructing a scattering characteristic matrix of the surface medium to radar waves in the target scanning area based on the backscatter coefficients at each time phase can be implemented using the following steps:

[0069] In step S1031, the backscatter coefficients at each time phase are obtained, and each pixel point in the target scanning area corresponds to a backscatter coefficient;

[0070] In step S1032, the backscatter coefficients at each time phase are arranged in time series to generate a backscatter coefficient vector at each pixel point in the target scanning area;

[0071] In step S1033, the backscatter coefficient vectors at all pixel points are combined into a two-dimensional matrix according to the spatial dimension to obtain the scattering characteristic matrix of the surface medium in the target scanning area to the radar wave.

[0072] In the specific implementation, first, the backscattering coefficients in each time phase are obtained, and each pixel point in the target scanning area corresponds to a backscattering coefficient; then, the backscattering coefficients corresponding to the same pixel point position in each time phase are arranged in time series to generate a backscattering coefficient vector at each pixel point in the target scanning area; finally, the backscattering coefficient vectors at all pixel points are combined into a two-dimensional matrix according to the spatial dimension, in which the row vector represents the spatial position of the pixel point and the column vector represents the backscattering coefficients in different time phases. The combined two-dimensional matrix is ​​used as the scattering characteristic matrix of the surface medium in the target scanning area to the radar wave.

[0073] It should be noted that, in this embodiment, the backscatter coefficient vector represents a combination of multiple backscatter coefficients arranged in chronological order; the scattering characteristic matrix in this application represents a matrix used to measure the response change law of the surface medium to the radar wave in the time dimension. Specifically, the scattering characteristic matrix refers to a two-dimensional numerical matrix formed by organizing the backscatter coefficients of all pixel points in the target scanning area in multiple phases according to spatial position and time series. The scattering characteristic matrix can be used to analyze the dynamic evolution characteristics of the physical state of the surface and is an important basic data structure for subsequent surface deformation sensitivity identification and deformation zone division.

[0074] In some embodiments, determining the surface deformation sensitivity of each pixel point in the target scanning area based on the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient can be achieved by the following steps:

[0075] Determining the scattering fluctuation intensity corresponding to each pixel point based on the backscattering coefficient vector corresponding to each pixel point in the scattering characteristic matrix;

[0076] Obtain the coherence coefficient sequence of each pixel in all time intervals;

[0077] Determining the spatiotemporal variation characteristics corresponding to each pixel point according to the coherence coefficient sequence corresponding to each pixel point;

[0078] For each pixel point, the scattering fluctuation intensity corresponding to the pixel point is normalized and weighted fused with the spatiotemporal variation characteristics to obtain the surface deformation sensitivity of each pixel point in the target scanning area.

[0079] It should be noted that the surface deformation sensitivity in this application represents an indicator that measures the degree of sensitivity of each pixel position in the target scanning area to surface deformation response. The higher the surface deformation sensitivity, the greater the fluctuation in the scattering intensity of the pixel point at different observation phases and the more unstable the coherence, indicating that the corresponding ground object or surface area may have activity changes or deformation trends, and has a higher surface deformation risk or sensitivity. Therefore, by determining the sensitivity to the surface deformation, it is possible to effectively identify those surface areas that show obvious abnormalities or instability in radar wave reflection characteristics. These areas are often highly correlated with geological bodies (such as landslides) that actually have activity or potential instability risks.

[0080] In specific implementation, first, the scattering fluctuation intensity corresponding to each pixel point is determined based on the backscattering coefficient vector corresponding to each pixel point in the scattering characteristic matrix, that is: for each pixel point, the backscattering coefficient variance of the backscattering coefficient vector corresponding to the pixel point is used as the scattering fluctuation intensity corresponding to the pixel point, and then the scattering fluctuation intensity corresponding to each pixel point is obtained; secondly, the coherence coefficient sequence of each pixel point in all time phase intervals is obtained, and the coherence coefficient sequence represents a combination of coherence coefficients including multiple time phase intervals; then, the spatiotemporal variation characteristics corresponding to each pixel point are determined according to the coherence coefficient sequence corresponding to each pixel point, that is: for each pixel point, the minimum and maximum ratio of the coherence coefficient of the coherence coefficient sequence corresponding to the pixel point is used as the spatiotemporal variation characteristics corresponding to the pixel point, and then the spatiotemporal variation characteristics corresponding to each pixel point are obtained. spatiotemporal variation characteristics; finally, for each pixel point, the scattering fluctuation intensity corresponding to the pixel point and the spatiotemporal variation characteristics are normalized and weightedly fused to obtain the surface deformation sensitivity of each pixel point in the target scanning area, that is: for each pixel point, the scattering fluctuation intensity and the spatiotemporal variation characteristics of the pixel point are normalized to between 0 and 1 through minimum-maximum normalization, and then the normalized scattering fluctuation intensity and the spatiotemporal variation characteristics are weightedly fused, and the weighted fusion result is used as the surface deformation sensitivity at the pixel point, thereby obtaining the surface deformation sensitivity of each pixel point in the target scanning area, wherein the weights of the scattering fluctuation intensity and the spatiotemporal variation characteristics can be set according to actual needs. For example, in this embodiment, the weights of the scattering fluctuation intensity and the spatiotemporal variation characteristics can be set to 0.6 and 0.4, respectively, which are not limited here.

[0081] It should be noted that, in this embodiment, the scattering fluctuation intensity represents the degree to which the backscattering coefficient of the pixel position in the target scanning area changes over time. The scattering fluctuation intensity is used to measure the strength of the change in the radar wave reflection characteristics of the pixel position under different observation phases. By determining the scattering fluctuation intensity, the surface deformation and landform changes in the target scanning area can be identified; in this embodiment, the spatiotemporal variation characteristics represent the degree to which the coherence coefficient of the pixel position in the target scanning area changes over time. The spatiotemporal variation characteristics reflect the fluctuation characteristics of the coherence of the pixel position in the entire time series, which are often used to judge the stability of the surface or identify potential deformation areas. The larger the spatiotemporal variation characteristics, the more drastic the coherence change, indicating that there are larger scatterer structure changes, landform disturbances or deformation activities on the surface. In landslide detection, by determining the spatiotemporal variation characteristics, the spatiotemporal resolution and accuracy of deformation identification can be effectively enhanced.

[0082] In step S104, singular value decomposition is performed on the scattering characteristic matrix, and deformation sensitive areas and stability partitions in the target scanning area are divided according to the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point.

[0083] In some embodiments, performing singular value decomposition on the scattering characteristic matrix may be implemented by the following steps:

[0084] Decomposing the scattering characteristic matrix into a product form of a left singular vector matrix, a singular value diagonal matrix and a right singular vector matrix;

[0085] All the singular values ​​are arranged in descending order, and the left singular vector corresponding to the largest singular value is extracted as the main singular vector, thereby completing the singular value decomposition of the scattering characteristic matrix.

[0086] In specific implementation, first, the scattering characteristic matrix is ​​decomposed into the product form of a left singular vector matrix, a singular value diagonal matrix and a right singular vector matrix using the existing singular value decomposition method, which will not be repeated here; then, all singular values ​​are arranged in descending order, and the singular vector corresponding to the largest singular value is extracted, thereby completing the singular value decomposition of the scattering characteristic matrix.

[0087] It should be noted that in this embodiment, the main singular vector is the vector corresponding to the maximum singular value in the left singular vector matrix. The left singular vector corresponding to the maximum singular value represents the spatial pattern with the most significant changes in the entire data set. Among them, the left singular vector corresponds to the spatial dimension and is used to reveal the response of the surface area (pixel) to the main change trend, while the right singular vector corresponds to the change trend in the time dimension and is more used to analyze the commonalities of time series. It cannot be directly used to divide the landslide area in space. Therefore, the left singular vector can be used to identify deformation-sensitive areas.

[0088] In some embodiments, reference Figure 3 As shown in FIG. 1 , this figure is an exemplary flow chart for determining deformation-sensitive areas and stability zones according to some embodiments of the present application. In this embodiment, the deformation-sensitive areas and stability zones in the target scanning area are divided according to the singular value decomposition results of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point. The following steps can be used to implement this:

[0089] In step S1041, the main singular vectors are extracted from the singular value decomposition result of the scattering characteristic matrix;

[0090] In step S1042, a principal component projection value of each pixel is determined according to the principal singular vector and the backscattering coefficient vector of each pixel in the scattering characteristic matrix;

[0091] In step S1043, a characteristic response map of the surface deformation in the target scanning area is determined by using the principal component projection value corresponding to each pixel point and the surface deformation sensitivity;

[0092] In step S1044, the target scanning area is divided into deformation sensitive areas and stability zones based on the characteristic response map.

[0093] In this application, the deformation-sensitive area refers to the spatial area with high sensitivity to surface deformation in the target scanning area. The pixel points in this type of area usually show abnormal changing trends in the backscattering characteristics, and their coherence coefficients fluctuate significantly in time series, indicating that these surface positions are more sensitive to external disturbances (such as rainfall, groundwater changes or loosening of slope structures), and are prone to subtle deformations or even landslide precursors. By identifying deformation-sensitive areas, the area with potential geological disaster risks is identified in advance, providing data support for subsequent landslide risk assessment, thereby realizing the advancement and precision of disaster prevention and control work; the stability partition in this application refers to the area with low sensitivity to surface deformation in the target scanning area, which is usually manifested as a gentle change in backscattering characteristics and a stable coherence coefficient, indicating that the surface state of these areas is relatively stable and no significant deformation or disturbance has occurred. The role of the stability partition is to provide a reference for landslide risk identification, while excluding risk-free areas, reducing false alarms, and improving detection accuracy.

[0094] In a specific implementation, first, the principal singular vector is extracted from the singular value decomposition result of the scattering characteristic matrix; secondly, the principal component projection value of each pixel point is determined according to the principal singular vector in combination with the backscattering coefficient vector of each pixel point in the scattering characteristic matrix, that is: for each pixel point, the backscattering coefficient vector of the loudness point is convolved with the principal singular vector, and the convolution result is used as the principal component projection value of the pixel point, thereby obtaining the principal component projection value of each pixel point; further, the characteristic response map of the surface deformation in the target scanning area is determined by the principal component projection value corresponding to each pixel point and the surface deformation sensitivity, that is: for each pixel point, the principal component projection value of the pixel point is used as a weight value to weight the corresponding surface deformation sensitivity, and the weighted result is used as the target scanning area. The characteristic response value of the surface deformation at the pixel point in the target scanning area is obtained, and then the characteristic response value of the surface deformation at each pixel point in the target scanning area is obtained, and all the characteristic response values ​​are arranged according to the spatial dimension of the target scanning area to obtain a characteristic response map of the surface deformation in the target scanning area; finally, the deformation sensitive area and the stability partition in the target scanning area are divided based on the characteristic response map, that is: the characteristic response value corresponding to each pixel point in the characteristic response map is compared with the characteristic response threshold, and the pixel point position with a characteristic response value less than the characteristic response threshold is classified as the stability zone in the target scanning area, and the pixel point position with a characteristic response value greater than or equal to the characteristic response threshold is classified as the stability zone in the target scanning area, wherein the characteristic response threshold can be set according to actual needs and will not be repeated here.

[0095] It should be noted that, in this embodiment, the principal component projection value represents the projection result of the pixel point in the target scanning area on the principal singular vector, and the principal component projection value reflects the response intensity of the pixel point to the main change trend of the radar scattering characteristics; in this embodiment, the characteristic response map represents the distribution characteristics of the potential deformation of the surface of the target scanning area on the temporal and spatial scales, and is an important intermediate result for identifying deformation-sensitive areas and stability zoning.

[0096] In step S105, the scattering entropy of the pixel points in the deformation sensitive area is determined, and the hidden landslide points in the target scanning area are identified from the stability partition according to the scattering entropy, and then a landslide risk probability map is generated by combining all the hidden landslide points with the deformation sensitive area.

[0097] In some embodiments, determining the scattering entropy of the pixel points in the deformation sensitive area may be achieved by using the following steps:

[0098] Extract the backscattering coefficient sequence of each pixel in the deformation sensitive area;

[0099] A pixel point is selected as a selected pixel point, and the scattering entropy of the selected pixel point is determined according to a backscattering coefficient sequence of the selected pixel point;

[0100] Continue to determine the scattering entropy of the remaining pixel points in the deformation sensitive area.

[0101] In a specific implementation, first, a backscattering coefficient sequence of each pixel point in the deformation-sensitive area is extracted, and the backscattering coefficient sequence includes backscattering coefficients of different phases; second, the scattering entropy of the selected pixel point is determined according to the backscattering coefficient sequence of the selected pixel point, that is, the probability distribution of different backscattering coefficients in the backscattering coefficient sequence is extracted, and the probability distribution is the frequency of occurrence of the backscattering coefficient in the backscattering coefficient sequence, and each extracted probability distribution is input as an input variable into the entropy model, and the entropy model outputs the scattering entropy of the selected pixel point. The entropy model adopts Shannon entropy, which will not be repeated here; finally, the scattering entropy of the remaining pixel points in the deformation-sensitive area is further determined by the determination method of "determining the scattering entropy of the selected pixel point according to the backscattering coefficient sequence of the selected pixel point".

[0102] It should be noted that the scattering entropy in this application represents the complexity of the fluctuation of the radar wave scattering intensity at the pixel position in the deformation-sensitive area. The scattering entropy is used to determine whether there are hidden landslide points in the stability partition that appear stable but actually have extremely complex scattering characteristics. Although these points do not have obvious deformation trends, they have unstable scattering characteristics due to underground structure disturbances, vegetation changes or hydrological disturbances, and have potential landslide risks. Landslide precursors are usually accompanied by slight surface disturbances and physical structure changes. Although they do not cause obvious deformation, they are sufficient to trigger radar scattering fluctuations. Therefore, the scattering entropy can be used to further identify hidden landslide points in the stability partition.

[0103] In some embodiments, identifying hidden landslide points in the target scanning area from the stability partitions according to the scattering entropy can be achieved by using the following steps:

[0104] determining a plurality of hidden landslide guiding points in a target scanning area according to the scattering entropy of each pixel point in the deformation sensitive area;

[0105] Extracting pixel points adjacent to each hidden landslide guiding point from the stability partition as candidate hidden landslide points;

[0106] Conduct neighborhood scattering deviation analysis on each candidate hidden landslide point to obtain the structural instability of each candidate hidden landslide point;

[0107] Multiple hidden landslide points in the target scanning area are extracted from all candidate hidden landslide points according to the degree of structural instability.

[0108] It should be noted that the hidden landslide points in this application refer to the location points in the target scanning area that do not show obvious deformation trends. These hidden landslide points are usually located in the stability zone. Because the surface deformation amplitude is small or there is no continuous change in the short term, they cannot be directly identified by conventional deformation sensitivity. Although these hidden landslide points have not developed into large-scale deformation, they are potential precursors to landslide formation. Therefore, by determining the hidden landslide points, the detection effectiveness of the landslide area can be effectively improved.

[0109] In the specific implementation, first, multiple hidden landslide guiding points in the target scanning area are determined according to the scattering entropy of each pixel point in the deformation sensitive area, that is, the scattering entropy of each pixel point is compared with the scattering entropy threshold, and the pixel points corresponding to the scattering entropy greater than the scattering entropy threshold are extracted as the hidden landslide guiding points in the target scanning area, thereby obtaining multiple hidden landslide guiding points in the target scanning area. The scattering entropy threshold can be set according to actual needs and is not limited here; secondly, the pixel points adjacent to each hidden landslide guiding point are extracted from the stability partition as candidate hidden landslide points; then, the neighborhood scattering deviation analysis is performed on each candidate hidden landslide point to obtain the structural instability of each candidate hidden landslide point, that is, for each hidden landslide point, the scattering entropy threshold is ... scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is not limited here; secondly, the scattering entropy threshold is set according to actual needs and is candidate hidden landslide points, extract pixel points within the neighborhood range of the candidate hidden landslide points as reference pixel points, calculate the mean deviation of the backscatter coefficient between the candidate hidden landslide points and all reference pixel points as the structural instability of the candidate hidden landslide points, and then obtain the structural instability of each candidate hidden landslide point. Among them, the neighborhood range adopts eight-domain range, which will not be repeated here; finally, compare the structural instability of each candidate hidden landslide point with the instability threshold, and take the candidate hidden landslide point with a structural instability greater than the structural instability threshold as the hidden landslide point of the target scanning area, and then obtain multiple hidden landslide points in the target scanning area. The instability threshold can be set according to actual needs or according to expert knowledge, and is not limited here.

[0110] It should be noted that, in this embodiment, the hidden landslide guiding point refers to an auxiliary point used to guide the identification of the hidden landslide point; in this embodiment, the candidate hidden landslide point refers to the initially identified hidden landslide point; in this embodiment, the structural instability refers to an indicator for measuring whether the radar scattering behavior of the candidate hidden landslide point in its local neighborhood in the stability partition significantly deviates from the statistical characteristics of its neighboring pixels. The structural instability reflects the instability or disturbance signs that may exist in the surface or subsurface structure of the candidate hidden landslide point. When landslide monitoring is performed in the target scanning area, the point with a larger scattering entropy indicates a stronger surface instability. When the surface instability is stronger, the surrounding area will generate hidden landslide risks. Therefore, the hidden landslide risk area can be identified by the scattering entropy.

[0111] In some embodiments, generating a landslide risk probability map by combining all hidden landslide points with the deformation sensitive area can be achieved by the following steps:

[0112] Combining the area formed by all hidden landslide points and the deformation sensitive area into a potential landslide risk area of ​​the target scanning area;

[0113] Extracting terrain gradient parameters at various locations in the potential landslide risk area, including slope, aspect, and elevation change rate;

[0114] Normalizing and fusing the terrain gradient parameters at each location in the potential landslide risk area to generate a landslide risk factor at each location;

[0115] The Kriging interpolation algorithm is used to expand the landslide risk factors at each location to the entire target scanning area to generate a landslide risk probability map.

[0116] In the specific implementation, first, all the hidden landslide points identified from the stability partition are integrated with the areas with high deformation sensitivity to form a potential landslide risk area. The potential landslide risk area reflects the spatiotemporal anomaly characteristics of the surface in electromagnetic scattering and coherence; secondly, the terrain gradient parameters of each location are extracted from the potential landslide risk area. The terrain gradient parameters include slope, slope aspect and elevation change rate. These terrain gradient parameters are obtained by analyzing the existing digital elevation model (DEM) data and will not be repeated here; then, for each location, the various parameters in the terrain gradient parameters are normalized and weightedly fused according to empirical weights to generate a landslide risk factor for each location; finally, the existing Kriging interpolation algorithm is used to smoothly extrapolate the risk factors of known locations to the entire target scanning area based on spatial autocorrelation to generate a landslide risk probability map, which will not be repeated here.

[0117] It should be noted that, in this embodiment, the potential landslide risk area refers to the spatial area in the target scanning area that shows a higher possibility of landslide occurrence; in this embodiment, the terrain gradient parameter refers to a quantitative indicator reflecting the trend of surface morphological changes and spatial undulation characteristics, which is used for geological disaster monitoring; in this embodiment, the landslide risk factor refers to an indicator used to characterize the possibility of landslide disasters. The landslide risk factor can reflect the regional surface stability, external inducing conditions and geological structure vulnerability, and is the core parameter for landslide risk modeling and probability map generation; the landslide risk probability map in this application represents the spatial distribution map of the possibility of landslide disasters at various locations in the target scanning area. By determining the landslide risk probability map, a visual early warning of landslide risks can be achieved, which assists in the scientific formulation of disaster prevention and control and land use strategies, and provides an intuitive, quantitative and dynamically updated decision-making basis for landslide monitoring, regional management and emergency response.

[0118] In addition, in another aspect of the present application, in some embodiments, the present application provides a landslide area detection system based on radar remote sensing data, referring to Figure 4 This figure is a schematic structural diagram of a landslide area detection system based on radar remote sensing data according to some embodiments of the present application. The landslide area detection system based on radar remote sensing data 200 includes: an acquisition module 201, a processing module 202, and an execution module 203, which are described as follows:

[0119] Acquisition module 201, in this application, acquisition module 201 is mainly used to acquire radar remote sensing data of the target scanning area in multiple time phases;

[0120] Processing module 202, in this application, the processing module 202 is mainly used to extract the backscatter coefficient and coherence coefficient when the detection radar detects the target scanning area in each phase from the corresponding radar remote sensing data;

[0121] The processing module 202 is further configured to construct a scattering characteristic matrix of the surface medium in the target scanning area to the radar wave based on the backscattering coefficients in each time phase, and determine the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient;

[0122] In addition, the processing module 202 is further configured to perform singular value decomposition on the scattering characteristic matrix, and divide the target scanning area into deformation sensitive areas and stability zones based on the singular value decomposition results of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point;

[0123] Execution module 203, in this application, execution module 203 is mainly used to determine the scattering entropy of the pixel points in the deformation sensitive area, identify the hidden landslide points in the target scanning area from the stability partition based on the scattering entropy, and then generate a landslide risk probability map by combining all the hidden landslide points with the deformation sensitive area.

[0124] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned landslide area detection method based on radar remote sensing data.

[0125] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device for implementing a landslide area detection method based on radar remote sensing data according to some embodiments of the present application. The landslide area detection method based on radar remote sensing data in the above embodiment can be Figure 5The computer device 300 shown in FIG. 1 is implemented as shown in FIG. 1 , and the computer device 300 includes at least one processor 301 , a communication bus 302 , a memory 303 , and at least one communication interface 304 .

[0126] The processor 301 may be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC) or one or more processors for controlling the execution of the landslide area detection method based on radar remote sensing data in the present application.

[0127] The communication bus 302 may be used to transmit information between the aforementioned components.

[0128] Memory 303 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer. Memory 303 may be independent and connected to processor 301 via communication bus 302. Memory 303 may also be integrated with processor 301.

[0129] Memory 303 is used to store program code for executing the solution of the present application, and is controlled by processor 301 for execution. Processor 301 is used to execute the program code stored in memory 303. The program code may include one or more software modules. The determination of the landslide area detection method based on radar remote sensing data in the above embodiment can be implemented by processor 301 and one or more software modules in the program code in memory 303.

[0130] The communication interface 304 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0131] In a specific implementation, as an example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0132] The aforementioned computer device can be a general-purpose computer device or a dedicated computer device. In a specific implementation, the computer device can be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of this application do not limit the type of computer device.

[0133] In addition, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned landslide area detection method based on radar remote sensing data.

[0134] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0135] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A landslide area detection method based on radar remote sensing data, characterized in that: The steps include: Acquire radar remote sensing data of the target scanning area in multiple time phases; Extract the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area in each phase from the corresponding radar remote sensing data; Based on the backscatter coefficients at each time phase, a scattering characteristic matrix of the surface medium in the target scanning area to radar waves is constructed. The surface deformation sensitivity of each pixel point in the target scanning area is determined based on the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient. For each pixel point, the minimum to maximum ratio of the coherence coefficient sequence corresponding to the pixel point is used as the spatiotemporal variation characteristic corresponding to the pixel point, thereby obtaining the spatiotemporal variation characteristic corresponding to each pixel point. Performing singular value decomposition on the scattering characteristic matrix, and dividing the target scanning area into deformation sensitive areas and stability zones according to the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point; The scattering entropy of the pixel points in the deformation sensitive area is determined, and the hidden landslide points in the target scanning area are identified from the stability partition according to the scattering entropy, and then a landslide risk probability map is generated by combining all the hidden landslide points with the deformation sensitive area.

2. The method according to claim 1, wherein The backscatter coefficient and coherence coefficient extracted from the corresponding radar remote sensing data at each time phase when the detection radar detects the target scanning area specifically include: Preprocess the radar remote sensing data at each time phase to generate single-view complex radar images of the target scanning area at each time phase; A time phase is selected as a selected time phase, and the backscatter coefficient of the detection radar detecting the target scanning area in the selected time phase is extracted from the single-look complex radar image corresponding to the selected time phase according to the radiation calibration function; Registering the single-view complex radar image corresponding to the selected time with the single-view complex radar image corresponding to the adjacent time to generate a single-view complex radar image pair corresponding to the selected time; Determining, based on the single-view complex radar image pair, a coherence coefficient when the detection radar detects the target scanning area at a selected time phase; Continue to determine the backscatter coefficient and coherence coefficient when the detection radar detects the target scanning area in the remaining time phase.

3. The method according to claim 1, wherein The scattering characteristic matrix of the surface medium to the radar wave in the target scanning area is constructed based on the backscattering coefficients in each phase. Specifically, it includes: Obtain the backscatter coefficient at each time phase. Different pixel points in the target scanning area correspond to a backscatter coefficient. Arrange the backscatter coefficients at each time phase in time series to generate a backscatter coefficient vector at each pixel point in the target scanning area; The backscattering coefficient vectors at all pixel points are combined into a two-dimensional matrix according to the spatial dimension to obtain the scattering characteristic matrix of the surface medium to the radar wave in the target scanning area.

4. The method according to claim 1, wherein Determining the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient specifically includes: Determining the scattering fluctuation intensity corresponding to each pixel point based on the backscattering coefficient vector corresponding to each pixel point in the scattering characteristic matrix; Obtain the coherence coefficient sequence of each pixel in all time intervals; Determining the spatiotemporal variation characteristics corresponding to each pixel point according to the coherence coefficient sequence corresponding to each pixel point; For each pixel point, the scattering fluctuation intensity corresponding to the pixel point is normalized and weighted fused with the spatiotemporal variation characteristics to obtain the surface deformation sensitivity of each pixel point in the target scanning area.

5. The method according to claim 1, wherein Decomposing the target scanning area into deformation-sensitive areas and stability zones based on the singular value decomposition results of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point specifically includes: extracting principal singular vectors from a singular value decomposition result of the scattering characteristic matrix; Determine the principal component projection value of each pixel point according to the principal singular vector combined with the backscattering coefficient vector of each pixel point in the scattering characteristic matrix; The characteristic response map of the surface deformation in the target scanning area is determined by the principal component projection value corresponding to each pixel point and the sensitivity of the surface deformation; The target scanning area is divided into deformation sensitive areas and stability zones based on the characteristic response map.

6. The method according to claim 1, wherein Determining the scattering entropy of the pixel points in the deformation sensitive area specifically includes: Extract the backscattering coefficient sequence of each pixel in the deformation sensitive area; A pixel point is selected as a selected pixel point, and the scattering entropy of the selected pixel point is determined according to a backscattering coefficient sequence of the selected pixel point; Continue to determine the scattering entropy of the remaining pixel points in the deformation sensitive area.

7. The method according to claim 1, wherein The radar remote sensing data of the target scanning area in multiple time phases is obtained through the detection radar.

8. A landslide area detection system based on radar remote sensing data, used to execute the landslide area detection method based on radar remote sensing data according to any one of claims 1 to 7, characterized in that: include: An acquisition module is used to acquire radar remote sensing data of the target scanning area in multiple time phases; A processing module is used to extract the backscatter coefficient and coherence coefficient of the detection radar when detecting the target scanning area in each time phase from the corresponding radar remote sensing data; The processing module is further configured to construct a scattering characteristic matrix of the surface medium in the target scanning area to the radar wave based on the backscattering coefficients in each time phase, and determine the surface deformation sensitivity of each pixel point in the target scanning area according to the scattering characteristic matrix combined with the spatiotemporal variation characteristics of the coherence coefficient; The processing module is further configured to perform singular value decomposition on the scattering characteristic matrix, and divide the target scanning area into deformation-sensitive areas and stability zones based on the singular value decomposition result of the scattering characteristic matrix and the surface deformation sensitivity of each pixel point; An execution module is used to determine the scattering entropy of the pixel points in the deformation sensitive area, identify the hidden landslide points in the target scanning area from the stability partition according to the scattering entropy, and then generate a landslide risk probability map by combining all the hidden landslide points with the deformation sensitive area.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute the landslide area detection method based on radar remote sensing data according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the landslide area detection method based on radar remote sensing data according to any one of claims 1 to 7 is implemented.

Images