Landslide identification method and system based on remote sensing monitoring

Abstract

The present invention provides a landslide identification method and system based on remote sensing monitoring, which remotely senses the entire target area, identifies the captured remote sensing images in a zoned manner, determines the remote sensing sub-images in which landslide terrain exists, and then uses the geological state information of the landslide terrain as a reference to judge whether a mud dumping event occurs and obtain mud dumping movement information, so that it can quickly and timely predict whether a predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

Description

Technical Field

[0001] The invention relates to the technical field of geological monitoring, and in particular to a landslide identification method and system based on remote sensing monitoring. Background Art

[0002] In mountainous areas, especially those with low vegetation coverage, water and soil erosion is prone to occur, which increases the geological disaster of mudslides in mountainous areas. At present, manual surveys are usually carried out in the corresponding mountainous areas to obtain the distribution of landslides in the mountainous areas. However, the manual survey method cannot effectively and comprehensively cover the entire mountainous area. At the same time, the field survey requires further processing of the obtained mountain geological data, which makes the calculation of the survey results have a certain time delay, thereby increasing the manpower and material resources required for landslide surveys, and there is also the problem of missing surveys. Summary of the invention

[0003] In view of the defects of the prior art, the present invention provides a landslide identification method and system based on remote sensing monitoring, which cuts a remote sensing image of a target area into a plurality of remote sensing sub-images, and divides each remote sensing sub-image into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain; extracts geological state information of the landslide terrain from the remote sensing sub-image containing the landslide terrain, thereby judging whether a mud dumping event will occur in the landslide terrain, and predicting corresponding mud dumping movement information; then, according to the mud dumping movement information, determines whether the mud dumping process will reach the location of a predetermined target object in the target area, and performs corresponding early warning operations, wherein the target area is remotely sensed as a whole, and the captured remote sensing images are partitioned for identification, and the remote sensing sub-images containing the landslide terrain are determined, and then, based on the geological state information of the landslide terrain, it is judged whether a mud dumping event will occur and obtains mud dumping movement information, so that it can quickly and timely predict whether a predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

[0004] The present invention provides a landslide identification method based on remote sensing monitoring, which comprises the following steps:

[0005] Step S1, acquiring a remote sensing image of a target area, performing cropping processing on the remote sensing image to obtain a plurality of remote sensing sub-images; performing recognition processing on each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain;

[0006] Step S2, analyzing and processing the remote sensing sub-image containing the landslide terrain to obtain geological status information corresponding to the landslide terrain; judging whether a mud dump event will occur in the landslide terrain according to the geological status information, and predicting the corresponding mud dump movement information when the mud dump event occurs;

[0007] Step S3, determining whether the mud pouring process will reach the location of the predetermined target object in the target area according to the mud pouring movement information; and performing corresponding early warning operations according to the determination result.

[0008] Further, in the step S1, a remote sensing image of the target area is obtained, and the remote sensing image is cropped to obtain a plurality of remote sensing sub-images; each remote sensing sub-image is identified and distinguished into remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain. Specifically, the remote sensing sub-images include:

[0009] Performing remote sensing scanning and photographing on the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour line distribution information of the target area from the panoramic remote sensing image, and cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour lines according to the contour line distribution information; wherein the contour line area refers to the area between two adjacent contour lines;

[0010] Pixel texture feature information contained in each remote sensing sub-image is extracted, and each remote sensing sub-image is divided into a remote sensing sub-image containing landslide terrain and a remote sensing sub-image not containing landslide terrain according to the pixel texture feature information.

[0011] Furthermore, in step S2, the remote sensing sub-image containing the landslide terrain is analyzed and processed to obtain geological state information corresponding to the landslide terrain; based on the geological state information, it is determined whether a mud dumping event will occur in the landslide terrain, and the corresponding mud dumping movement information when the mud dumping event occurs is predicted, which specifically includes:

[0012] According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain;

[0013] Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude of the gravity force on each part of the landslide terrain along the slope surface;

[0014] According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event;

[0015] According to the land area information and the landslide inclination information, the corresponding mud pouring movement path and the mud pouring movement average speed when the mud pouring event occurs are predicted as the mud pouring movement information.

[0016] Further, in the step S3, according to the mud pouring movement information, it is determined whether the mud pouring process will reach the location of the predetermined target object in the target area; according to the determination result, performing corresponding early warning operations specifically includes:

[0017] Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area;

[0018] If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; and based on the required time, a voice broadcast warning message including the required time is generated.

[0019] The present invention also provides a landslide identification system based on remote sensing monitoring, which comprises:

[0020] A remote sensing image acquisition and processing module is used to obtain a remote sensing image of a target area, crop the remote sensing image to obtain a plurality of remote sensing sub-images; identify each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain;

[0021] A landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological status information of the corresponding landslide terrain;

[0022] A mud dump event analysis module, which is used to determine whether a mud dump event will occur in the landslide terrain according to the geological state information, and to predict the corresponding mud dump movement information when a mud dump event occurs;

[0023] The mud pouring warning module is used to determine whether the mud pouring process will reach the location of the predetermined target object in the target area according to the mud pouring movement information; and perform corresponding warning operations according to the determination result.

[0024] Furthermore, the remote sensing image acquisition and processing module acquires the remote sensing image of the target area multiple times when acquiring the remote sensing image of the target area, and performs superposition optimization processing on each pixel point in the remote sensing image according to the remote sensing images of the target area acquired multiple times, so as to obtain the final remote sensing image of the target area, and the process is as follows:

[0025] Step A1, using the following formula (1), an overall image matrix analysis is performed based on the remote sensing images of the target area collected multiple times, so as to screen out some unqualified remote sensing images of the target area.

[0026]

[0027] In the above formula (1), Q(a) represents the screening value of the remote sensing image of the a-th target area; H a represents the image matrix of the remote sensing image of the a-th target area; H b The image matrix representing the remote sensing image of the b-th target area; |H a -H b | means to form a new matrix by subtracting the corresponding elements in the two image matrices and taking their absolute values; sum() means to sum all the elements in the image matrix; n means the total number of remote sensing images of the target area collected;

[0028] If Q(a) = 1, the remote sensing image of the ath target area is eliminated;

[0029] If Q(a) = 0, the remote sensing image of the a-th target area is retained;

[0030] The remote sensing image of the target area corresponding to Q(a) = 0 is selected and recorded as h k An image matrix representing the remote sensing image of the remaining k-th qualified target area;

[0031] Step A2, using the following formula (2), select the best remote sensing image of the target area from the remaining qualified remote sensing images of the target area,

[0032]

[0033] In the above formula (2), K indicates that the optimal remote sensing image of the target area is the remote sensing image of the remaining Kth qualified target area; G indicates the total number of remote sensing images of the target area remaining after screening in step A1; h k (i, j) represents the pixel value of the i-th row and j-th column in the image matrix of the remote sensing image of the remaining k-th qualified target area; I represents the total number of pixels in any column in the image matrix of the remote sensing image; J represents the total number of pixels in any row in the image matrix of the remote sensing image; It means that the value of k from 1 to G is substituted into the brackets to obtain the k value when the minimum value is obtained in the brackets;

[0034] Step A3, using the following formula (3), according to the difference between the remote sensing images of the remaining qualified target areas and the remote sensing images of the optimal target areas, each pixel in the remote sensing images of the optimal target areas is subjected to superposition optimization processing to obtain the final remote sensing images of the target areas.

[0035]

[0036] In the above formula (3), h(i,j) represents the pixel value at the i-th row and j-th column in the image matrix of the remote sensing image of the final target area after the overlay optimization process; h K (i, j) represents the pixel value at the i-th row and j-th column position in the image matrix of the remote sensing image of the remaining K-th qualified target area.

[0037] Further, the remote sensing image acquisition and processing module acquires the remote sensing image of the target area, performs cropping processing on the remote sensing image to obtain a plurality of remote sensing sub-images; performs recognition processing on each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain, specifically including:

[0038] Performing remote sensing scanning and photographing on the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour line distribution information of the target area from the panoramic remote sensing image, and cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour lines according to the contour line distribution information; wherein the contour line area refers to the area between two adjacent contour lines;

[0039] Pixel texture feature information contained in each remote sensing sub-image is extracted, and each remote sensing sub-image is divided into a remote sensing sub-image containing landslide terrain and a remote sensing sub-image not containing landslide terrain according to the pixel texture feature information.

[0040] Furthermore, the landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological status information corresponding to the landslide terrain, which specifically includes:

[0041] According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain;

[0042] The mud dumping event analysis module determines whether a mud dumping event will occur in the landslide terrain according to the geological state information, and predicts the corresponding mud dumping movement information when a mud dumping event occurs, specifically including:

[0043] Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude of the gravity force on each part of the landslide terrain along the slope surface;

[0044] According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event;

[0045] According to the land area information and the landslide inclination information, the corresponding mud pouring movement path and the mud pouring movement average speed when the mud pouring event occurs are predicted as the mud pouring movement information.

[0046] Further, the mud pouring warning module determines whether the mud will reach the location of the predetermined target object in the target area during the mud pouring process according to the mud pouring movement information; and according to the determination result, performing corresponding warning operations specifically includes:

[0047] Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area;

[0048] If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; and based on the required time, a voice broadcast warning message including the required time is generated.

[0049] Compared with the prior art, the landslide identification method and system based on remote sensing monitoring cuts the remote sensing image of the target area into several remote sensing sub-images, and divides each remote sensing sub-image into a remote sensing sub-image containing the landslide terrain and a remote sensing sub-image not containing the landslide terrain; extracts the geological state information of the landslide terrain from the remote sensing sub-image containing the landslide terrain, thereby judging whether a mud dumping event will occur in the landslide terrain, and predicting the corresponding mud dumping movement information; then, according to the mud dumping movement information, determines whether the mud dumping process will reach the location of the predetermined target object in the target area, and performs corresponding early warning operations, and remote sensing photography of the target area as a whole is performed, and the photographed remote sensing images are partitioned to identify the remote sensing sub-images containing the landslide terrain, and then, based on the geological state information of the landslide terrain, judges whether a mud dumping event occurs and obtains the mud dumping movement information, so that it can quickly and timely predict whether the predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

[0050] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0051] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A schematic flow chart of a landslide identification method based on remote sensing monitoring provided by the present invention.

[0054] Figure 2 This is a schematic diagram of the structure of the landslide identification system based on remote sensing monitoring provided by the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1 , is a flow chart of a landslide identification method based on remote sensing monitoring provided by an embodiment of the present invention. The landslide identification method based on remote sensing monitoring comprises the following steps:

[0057] Step S1, obtaining a remote sensing image of a target area, cutting the remote sensing image to obtain a plurality of remote sensing sub-images; performing recognition processing on each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain;

[0058] Step S2, analyzing and processing the remote sensing sub-image containing the landslide terrain to obtain geological status information corresponding to the landslide terrain; judging whether a mud dump event will occur in the landslide terrain according to the geological status information, and predicting the corresponding mud dump movement information when the mud dump event occurs;

[0059] Step S3, determining whether the mud pouring process will reach the location of the predetermined target object in the target area according to the mud pouring movement information; and performing corresponding early warning operations according to the determination result.

[0060] The beneficial effects of the above technical scheme are as follows: the landslide identification method based on remote sensing monitoring cuts the remote sensing image of the target area into several remote sensing sub-images, and divides each remote sensing sub-image into a remote sensing sub-image containing the landslide terrain and a remote sensing sub-image not containing the landslide terrain; extracts the geological state information of the landslide terrain from the remote sensing sub-image containing the landslide terrain, thereby judging whether a mud dumping event will occur in the landslide terrain, and predicting the corresponding mud dumping movement information; then, according to the mud dumping movement information, determines whether the mud dumping process will reach the location of the predetermined target object in the target area, and performs corresponding early warning operations, wherein the target area is remotely sensed as a whole, and the captured remote sensing images are partitioned to identify the remote sensing sub-images containing the landslide terrain, and then, based on the geological state information of the landslide terrain, judges whether a mud dumping event occurs and obtains the mud dumping movement information, so that it can quickly and timely predict whether the predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

[0061] Preferably, in step S1, a remote sensing image of the target area is acquired, and the remote sensing image is cropped to obtain a plurality of remote sensing sub-images; each remote sensing sub-image is identified and distinguished into remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain, which specifically includes:

[0062] Performing remote sensing scanning and photographing of the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour line distribution information of the target area from the panoramic remote sensing image, and cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour lines according to the contour line distribution information; wherein the contour line area refers to the area between two adjacent contour lines;

[0063] The pixel texture feature information contained in each remote sensing sub-image is extracted, and according to the pixel texture feature information, each remote sensing sub-image is divided into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain.

[0064] The beneficial effects of the above technical solution are as follows: in actual work, the remote sensing satellite is instructed to perform remote sensing scanning and shooting of the target area to obtain a panoramic remote sensing image of the entire target area, and then the contour distribution of the target area is calibrated from the panoramic remote sensing image, so that a number of contour lines in the panoramic remote sensing image divide it into a number of different contour areas, and then each contour area is used as a reference to cut the panoramic remote sensing image into a number of remote sensing sub-images corresponding to each contour area, so as to facilitate the detailed analysis of the landslide terrain for each remote sensing sub-image. In addition, the pixel texture distribution density of the landslide terrain and the non-landslide terrain in the corresponding part of the remote sensing sub-image is not the same. By identifying the pixel texture distribution density of the remote sensing sub-image, it is possible to determine whether the remote sensing sub-image includes the landslide terrain, thereby accurately distinguishing the remote sensing sub-image.

[0065] Preferably, in step S2, the remote sensing sub-image containing the landslide terrain is analyzed and processed to obtain geological state information corresponding to the landslide terrain; based on the geological state information, it is determined whether a mud dumping event will occur in the landslide terrain, and the corresponding mud dumping movement information when the mud dumping event occurs is predicted, which specifically includes:

[0066] According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain;

[0067] Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude information of the gravity force acting on each part of the landslide terrain along the slope surface;

[0068] According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event;

[0069] According to the land area information and the landslide inclination information, the corresponding soil pouring movement path and the average speed of the soil pouring movement when the soil pouring event occurs are predicted as the soil pouring movement information.

[0070] The beneficial effects of the above technical solution are: extracting pixel contour information from the remote sensing sub-image containing the landslide terrain, taking the pixel contour information as a reference, determining the landslide terrain in the corresponding contour area (i.e., the landslide boundary range) and the landslide inclination angle, and determining the gravity force on each part of the landslide terrain along the slope surface based on the distribution of soil accumulation height in the landslide terrain and the geometric appearance information of the landslide terrain, so as to judge whether each part of the landslide terrain is in a force balance state (i.e., whether the gravity force on each part of the landslide terrain in all directions is balanced), thereby quantitatively judging whether a landslide terrain will have a mud dumping event, so as to effectively predict whether a mud dumping event will occur. In addition, according to the landslide terrain landslide area information and landslide inclination information, the corresponding mud dumping movement path and mud dumping movement average speed when a mud dumping event occurs can be more accurately predicted, which is convenient for the subsequent identification of the impact range of the mud dumping.

[0071] Preferably, in step S3, according to the soil pouring movement information, it is determined whether the soil will reach the location of the predetermined target object in the target area during the soil pouring process; and according to the determination result, performing corresponding early warning operations specifically includes:

[0072] Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area;

[0073] If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; based on the required time, a voice broadcast warning message including the required time is generated.

[0074] The beneficial effect of the above technical solution is: through the above method, it is possible to predict whether a mudslide event will affect predetermined target objects (such as buildings, etc.) in the target area when a mudslide event occurs, so as to generate corresponding voice broadcast warning messages in time, thereby improving the early warning of geological accidents.

[0075] See also Figure 2 , is a schematic diagram of the structure of a landslide identification system based on remote sensing monitoring provided by an embodiment of the present invention. The landslide identification system based on remote sensing monitoring includes:

[0076] The remote sensing image acquisition and processing module is used to obtain the remote sensing image of the target area, crop the remote sensing image to obtain a number of remote sensing sub-images; identify each remote sensing sub-image and distinguish it into a remote sensing sub-image containing landslide terrain and a remote sensing sub-image not containing landslide terrain;

[0077] A landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological status information of the corresponding landslide terrain;

[0078] A mud dump event analysis module is used to determine whether a mud dump event will occur in the landslide terrain according to the geological state information, and to predict the corresponding mud dump movement information when a mud dump event occurs;

[0079] The mud pouring warning module is used to determine whether the mud pouring process will reach the location of the predetermined target object in the target area according to the mud pouring movement information; and perform corresponding warning operations according to the determination result.

[0080] The beneficial effects of the above technical solution are as follows: the landslide identification system based on remote sensing monitoring cuts the remote sensing image of the target area into several remote sensing sub-images, and divides each remote sensing sub-image into a remote sensing sub-image containing the landslide terrain and a remote sensing sub-image not containing the landslide terrain; extracts the geological state information of the landslide terrain from the remote sensing sub-image containing the landslide terrain, thereby judging whether a mud dumping event will occur in the landslide terrain, and predicting the corresponding mud dumping movement information; then, according to the mud dumping movement information, determines whether the mud dumping process will reach the location of the predetermined target object in the target area, and performs corresponding early warning operations, and remote sensing photography of the target area as a whole is performed, and the captured remote sensing images are partitioned to identify the remote sensing sub-images containing the landslide terrain, and then, based on the geological state information of the landslide terrain, judges whether a mud dumping event occurs and obtains the mud dumping movement information, so that it can quickly and timely predict whether the predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

[0081] Preferably, the remote sensing image acquisition and processing module acquires the remote sensing image of the target area multiple times when acquiring the remote sensing image of the target area, and performs superposition optimization processing on each pixel point in the remote sensing image according to the remote sensing images of the target area acquired multiple times, so as to obtain the final remote sensing image of the target area, and the process is as follows:

[0082] Step A1, using the following formula (1), an overall image matrix analysis is performed based on the remote sensing images of the target area collected multiple times, so as to screen out some unqualified remote sensing images of the target area.

[0083]

[0084] In the above formula (1), Q(a) represents the screening value of the remote sensing image of the a-th target area; H a represents the image matrix of the remote sensing image of the a-th target area; H b The image matrix representing the remote sensing image of the b-th target area; |H a -H b | means to form a new matrix by subtracting the corresponding elements in the two image matrices and taking their absolute values; sum() means to sum all the elements in the image matrix; n means the total number of remote sensing images of the target area collected;

[0085] If Q(a) = 1, the remote sensing image of the ath target area is eliminated;

[0086] If Q(a) = 0, the remote sensing image of the a-th target area is retained;

[0087] The remote sensing image of the target area corresponding to Q(a) = 0 is selected and recorded as h k An image matrix representing the remote sensing image of the remaining k-th qualified target area;

[0088] Step A2, using the following formula (2), select the best remote sensing image of the target area from the remaining qualified remote sensing images of the target area,

[0089]

[0090] In the above formula (2), K indicates that the optimal remote sensing image of the target area is the remote sensing image of the remaining Kth qualified target area; G indicates the total number of remote sensing images of the target area remaining after screening in step A1; h k (i, j) represents the pixel value of the i-th row and j-th column in the image matrix of the remote sensing image of the remaining k-th qualified target area; I represents the total number of pixels in any column in the image matrix of the remote sensing image; J represents the total number of pixels in any row in the image matrix of the remote sensing image; It means that the value of k from 1 to G is substituted into the brackets to obtain the k value when the minimum value is obtained in the brackets;

[0091] Step A3, using the following formula (3), according to the difference between the remote sensing images of the remaining qualified target areas and the remote sensing images of the optimal target areas, each pixel in the remote sensing images of the optimal target areas is subjected to superposition optimization processing to obtain the final remote sensing images of the target areas.

[0092]

[0093] In the above formula (3), h(i,j) represents the pixel value at the i-th row and j-th column in the image matrix of the remote sensing image of the final target area after the overlay optimization process; h K (i, j) represents the pixel value at the i-th row and j-th column position in the image matrix of the remote sensing image of the remaining K-th qualified target area.

[0094] The beneficial effects of the above technical solution are as follows: using the above formula (1), an overall analysis of the image matrix is ​​performed based on the remote sensing images of the target area collected multiple times, so as to screen out some unqualified remote sensing images of the target area, thereby preferentially eliminating unqualified images to make subsequent calculations more efficient; then using the above formula (2), the optimal remote sensing image of the target area is selected from the remaining qualified remote sensing images of the target area, and then the optimal remote sensing image is first selected as the initial image and further optimized; finally, using the above formula (3), based on the difference between the remaining qualified remote sensing images of the target area and the optimal remote sensing image of the target area, each pixel in the remote sensing image of the optimal target area is superimposed and optimized to obtain the final remote sensing image of the target area, thereby ensuring the reliability of the system and improving the accuracy of image acquisition.

[0095] Preferably, the remote sensing image acquisition and processing module acquires a remote sensing image of the target area, performs cropping processing on the remote sensing image to obtain a plurality of remote sensing sub-images; performs recognition processing on each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain, specifically including:

[0096] Performing remote sensing scanning and photographing of the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour line distribution information of the target area from the panoramic remote sensing image, and cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour lines according to the contour line distribution information; wherein the contour line area refers to the area between two adjacent contour lines;

[0097] The pixel texture feature information contained in each remote sensing sub-image is extracted, and according to the pixel texture feature information, each remote sensing sub-image is divided into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain.

[0098] The beneficial effects of the above technical solution are as follows: in actual work, the remote sensing satellite is instructed to perform remote sensing scanning and shooting of the target area to obtain a panoramic remote sensing image of the entire target area, and then the contour distribution of the target area is calibrated from the panoramic remote sensing image, so that a number of contour lines in the panoramic remote sensing image divide it into a number of different contour areas, and then each contour area is used as a reference to cut the panoramic remote sensing image into a number of remote sensing sub-images corresponding to each contour area, so as to facilitate the detailed analysis of the landslide terrain for each remote sensing sub-image. In addition, the pixel texture distribution density of the landslide terrain and the non-landslide terrain in the corresponding part of the remote sensing sub-image is not the same. By identifying the pixel texture distribution density of the remote sensing sub-image, it is possible to determine whether the remote sensing sub-image includes the landslide terrain, thereby accurately distinguishing the remote sensing sub-image.

[0099] Preferably, the landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological state information corresponding to the landslide terrain, specifically including:

[0100] According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain;

[0101] The mud dump event analysis module determines whether a mud dump event will occur in the landslide terrain according to the geological state information, and predicts the corresponding mud dump movement information when a mud dump event occurs, specifically including:

[0102] Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude information of the gravity force acting on each part of the landslide terrain along the slope surface;

[0103] According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event;

[0104] According to the land area information and the landslide inclination information, the corresponding soil pouring movement path and the average speed of the soil pouring movement when the soil pouring event occurs are predicted as the soil pouring movement information.

[0105] The beneficial effects of the above technical solution are: extracting pixel contour information from the remote sensing sub-image containing the landslide terrain, taking the pixel contour information as a reference, determining the landslide terrain in the corresponding contour area (i.e., the landslide boundary range) and the landslide inclination angle, and determining the gravity force on each part of the landslide terrain along the slope surface based on the distribution of soil accumulation height in the landslide terrain and the geometric appearance information of the landslide terrain, so as to judge whether each part of the landslide terrain is in a force balance state (i.e., whether the gravity force on each part of the landslide terrain in all directions is balanced), thereby quantitatively judging whether a landslide terrain will have a mud dumping event, so as to effectively predict whether a mud dumping event will occur. In addition, according to the landslide terrain landslide area information and landslide inclination information, the corresponding mud dumping movement path and mud dumping movement average speed when a mud dumping event occurs can be more accurately predicted, which is convenient for the subsequent identification of the impact range of the mud dumping.

[0106] Preferably, the mud pouring warning module determines whether the mud pouring process will reach the location of the predetermined target object in the target area according to the mud pouring movement information; and according to the determination result, performing corresponding warning operations specifically includes:

[0107] Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area;

[0108] If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; based on the required time, a voice broadcast warning message including the required time is generated.

[0109] The beneficial effect of the above technical solution is: through the above method, it is possible to predict whether a mudslide event will affect predetermined target objects (such as buildings, etc.) in the target area when a mudslide event occurs, so as to generate corresponding voice broadcast warning messages in time, thereby improving the early warning of geological accidents.

[0110] As can be seen from the contents of the above embodiments, the landslide identification method and system based on remote sensing monitoring cuts the remote sensing image of the target area into several remote sensing sub-images, and divides each remote sensing sub-image into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain; extracts the geological state information of the landslide terrain from the remote sensing sub-image containing the landslide terrain, thereby judging whether a mud dumping event will occur in the landslide terrain, and predicting the corresponding mud dumping movement information; then, according to the mud dumping movement information, determines whether the mud dumping process will reach the location of the predetermined target object in the target area, and performs corresponding early warning operations, wherein the target area is remotely sensed as a whole, and the captured remote sensing images are partitioned to identify the remote sensing sub-images containing the landslide terrain, and then, based on the geological state information of the landslide terrain, judges whether a mud dumping event occurs and obtains the mud dumping movement information, so that it can quickly and timely predict whether the predetermined target object in the target area will be affected by the mud dumping, thereby reducing the manpower and material resources required for landslide exploration and improving the accuracy and reliability of landslide identification.

[0111] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A landslide identification method based on remote sensing monitoring, characterized in that: It includes the following steps: Step S1, acquiring a remote sensing image of a target area, performing cropping processing on the remote sensing image to obtain a plurality of remote sensing sub-images; performing recognition processing on each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain; Step S2, analyzing and processing the remote sensing sub-image containing the landslide terrain to obtain geological status information corresponding to the landslide terrain; judging whether a mud dump event will occur in the landslide terrain according to the geological status information, and predicting the corresponding mud dump movement information when the mud dump event occurs; Step S3, determining whether the soil will reach the location of a predetermined target object in the target area during the soil pouring process according to the soil pouring movement information; and performing corresponding early warning operations according to the determination result; Wherein, in the step S1, a remote sensing image of a target area is obtained, and the remote sensing image is cropped to obtain a plurality of remote sensing sub-images; each remote sensing sub-image is identified and distinguished into remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain, which specifically include: Performing remote sensing scanning and photographing on the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour distribution information of the target area from the panoramic remote sensing image, and according to the contour distribution information, cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour areas; wherein the contour area refers to the area between two adjacent contour lines; extracting pixel texture feature information contained in each remote sensing sub-image, and according to the pixel texture feature information, distinguishing each remote sensing sub-image into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain; Wherein, in the step S2, the remote sensing sub-image containing the landslide terrain is analyzed and processed to obtain the geological state information corresponding to the landslide terrain; based on the geological state information, it is judged whether a mud dumping event will occur in the landslide terrain, and the corresponding mud dumping movement information when the mud dumping event occurs is predicted, which specifically includes: According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain; Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude of the gravity force on each part of the landslide terrain along the slope surface; According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event; According to the land area information and the landslide inclination information, the corresponding mud pouring movement path and the mud pouring movement average speed when the mud pouring event occurs are predicted as the mud pouring movement information.

2. The landslide identification method based on remote sensing monitoring as claimed in claim 1, characterized in that: In the step S3, according to the soil pouring movement information, it is determined whether the soil will reach the location of the predetermined target object in the target area during the soil pouring process; According to the determination result, the corresponding early warning operation specifically includes: Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area; If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; and based on the required time, a voice broadcast warning message including the required time is generated.

3. The landslide identification system based on remote sensing monitoring is characterized by: It includes: A remote sensing image acquisition and processing module is used to obtain a remote sensing image of a target area, crop the remote sensing image to obtain a plurality of remote sensing sub-images; identify each remote sensing sub-image to distinguish between remote sensing sub-images containing landslide terrain and remote sensing sub-images not containing landslide terrain; A landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological status information of the corresponding landslide terrain; A mud dump event analysis module, which is used to determine whether a mud dump event will occur in the landslide terrain according to the geological state information, and to predict the corresponding mud dump movement information when a mud dump event occurs; A mud dump early warning module, which is used to determine whether the mud dump will reach the location of a predetermined target object in the target area according to the mud dump movement information; and perform corresponding early warning operations according to the determination result; The remote sensing image acquisition and processing module acquires the remote sensing image of the target area, performs cropping processing on the remote sensing image, and obtains a plurality of remote sensing sub-images; performs recognition processing on each remote sensing sub-image, and distinguishes the remote sensing sub-images containing the landslide terrain and the remote sensing sub-images not containing the landslide terrain, which specifically include: Performing remote sensing scanning and photographing on the target area to obtain a panoramic remote sensing image corresponding to the target area; identifying contour distribution information of the target area from the panoramic remote sensing image, and according to the contour distribution information, cutting the panoramic remote sensing image into a plurality of remote sensing sub-images corresponding to different contour areas; wherein the contour area refers to the area between two adjacent contour lines; extracting pixel texture feature information contained in each remote sensing sub-image, and according to the pixel texture feature information, distinguishing each remote sensing sub-image into a remote sensing sub-image containing a landslide terrain and a remote sensing sub-image not containing a landslide terrain; The landslide terrain recognition module is used to analyze and process the remote sensing sub-image containing the landslide terrain to obtain the geological status information corresponding to the landslide terrain, which specifically includes: According to the pixel contour information of the remote sensing sub-image containing the landslide terrain, the landslide terrain in the corresponding contour area and the landslide inclination information, as well as the soil accumulation height information of the landslide terrain are determined, which is used as the geological state information of the landslide terrain; The mud dumping event analysis module determines whether a mud dumping event will occur in the landslide terrain according to the geological state information, and predicts the corresponding mud dumping movement information when a mud dumping event occurs, specifically including: Determine the gravity force distribution information of the landslide terrain according to the landslide range information, the landslide inclination information and the soil accumulation height information; wherein the gravity force distribution information indicates the magnitude of the gravity force on each part of the landslide terrain along the slope surface; According to the gravity force distribution information, determine whether each part of the landslide terrain is in a force balance state, if so, judge that the landslide terrain will not cause a mud dumping event; if not, judge that the landslide terrain will cause a mud dumping event; According to the land area information and the landslide inclination information, the corresponding mud pouring movement path and the mud pouring movement average speed when the mud pouring event occurs are predicted as the mud pouring movement information.

4. The landslide identification system based on remote sensing monitoring as claimed in claim 3 is characterized in that: The remote sensing image acquisition and processing module acquires the remote sensing image of the target area multiple times when acquiring the remote sensing image of the target area, and performs superposition optimization processing on each pixel point in the remote sensing image according to the remote sensing images of the target area acquired multiple times, so as to obtain the final remote sensing image of the target area, and the process is as follows: Step A1, using the following formula (1), an overall image matrix analysis is performed based on the remote sensing images of the target area collected multiple times, so as to screen out some unqualified remote sensing images of the target area. In the above formula (1), Q(a) represents the screening value of the remote sensing image of the a-th target area; H a represents the image matrix of the remote sensing image of the a-th target area; H b The image matrix representing the remote sensing image of the b-th target area; |H a -H b | means to form a new matrix by subtracting the corresponding elements in the two image matrices and taking their absolute values; sum() means to sum all the elements in the image matrix; n means the total number of remote sensing images of the target area collected; If Q(a) = 1, the remote sensing image of the ath target area is eliminated; If Q(a) = 0, the remote sensing image of the a-th target area is retained; The remote sensing image of the target area corresponding to Q(a) = 0 is selected and recorded as h k An image matrix representing the remote sensing image of the remaining k-th qualified target area; Step A2, using the following formula (2), select the best remote sensing image of the target area from the remaining qualified remote sensing images of the target area, In the above formula (2), K indicates that the optimal remote sensing image of the target area is the remote sensing image of the remaining Kth qualified target area; G indicates the total number of remote sensing images of the target area remaining after screening in step A1; h k (i, j) represents the pixel value of the i-th row and j-th column in the image matrix of the remote sensing image of the remaining k-th qualified target area; I represents the total number of pixels in any column in the image matrix of the remote sensing image; J represents the total number of pixels in any row in the image matrix of the remote sensing image; It means that the value of k from 1 to G is substituted into the brackets to obtain the k value when the minimum value is obtained in the brackets; Step A3, using the following formula (3), according to the difference between the remote sensing images of the remaining qualified target areas and the remote sensing images of the optimal target areas, each pixel in the remote sensing images of the optimal target areas is subjected to superposition optimization processing to obtain the final remote sensing images of the target areas. In the above formula (3), h(i,j) represents the pixel value at the i-th row and j-th column in the image matrix of the remote sensing image of the final target area after the overlay optimization process; h K (i, j) represents the pixel value at the i-th row and j-th column position in the image matrix of the remote sensing image of the remaining K-th qualified target area.

5. The landslide identification system based on remote sensing monitoring as claimed in claim 4, characterized in that: The mud pouring warning module determines whether the mud will reach the location of the predetermined target object in the target area during the mud pouring process according to the mud pouring movement information; According to the determination result, the corresponding early warning operation specifically includes: Determine whether the position of the predetermined target object in the target area is located on the mud pouring movement path, if so, determine that the mud pouring process will reach the position of the predetermined target object in the target area; if not, determine that the mud pouring process will not reach the position of the predetermined target object in the target area; If the mud pouring process will reach the location of the predetermined target object in the target area, the time required for the mud pouring process to reach the location of the predetermined target object in the target area is estimated based on the average moving speed of the mud pouring; and based on the required time, a voice broadcast warning message including the required time is generated.

Images