A landslide deformation early warning method
Through laser point cloud separation and contour structure analysis, the problem of low landslide deformation warning accuracy in traditional landslide monitoring methods is solved, and a comprehensive analysis and accurate warning of multi-angle landslide deformation intensity is achieved.
Patent Information
- Application Number
- CN202411230577.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-04
AI Technical Summary
Traditional landslide monitoring methods have problems such as high equipment cost, limited monitoring range, complex data processing and low image analysis accuracy, which cannot accurately reflect the landslide deformation situation, resulting in low warning accuracy of landslide deformation.
Using laser point cloud acquisition and separation technology, the laser point cloud is separated into XY two-dimensional point cloud, XZ two-dimensional point cloud and YZ two-dimensional point cloud, and the contour structure is extracted according to the reflection intensity, and the morphological value and landslide deformation intensity are calculated to achieve A comprehensive analysis of landslide deformation from multiple surfaces and provide early warning when the landslide deformation intensity is greater than the threshold.
The accuracy of landslide deformation warning is improved, and the landslide deformation intensity is analyzed through multiple angles, the data volume is simplified, and the accuracy and reliability of the warning is enhanced.
Smart Images

Figure CN118729982B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of landslide deformation early warning, and in particular to a landslide deformation early warning method. Background Art
[0002] Landslide is a common form of geological disaster. It refers to the phenomenon that rocks, soil or other surface materials move downward along a certain sliding surface in whole or in part under the influence of natural or human factors. Landslides are often sudden and destructive, posing a serious threat to people's lives, property and infrastructure. Therefore, timely and accurate monitoring and early warning of landslide deformation are crucial.
[0003] Traditional landslide monitoring methods include ground measurement, GPS monitoring, inclinometers and other means. Although these methods can provide displacement information of landslides to a certain extent, they often have problems such as high equipment cost, limited monitoring range, and complex data processing. With the development of science and technology, image monitoring and processing technology has been widely used in topographic mapping and landslide monitoring due to its high precision, fast speed and non-contact characteristics. However, when using images to analyze landslide deformation, the image is two-dimensional data and cannot accurately reflect the landslide deformation. There is a problem of low accuracy in landslide deformation warning. Summary of the invention
[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a landslide deformation early warning method which solves the problem of low landslide deformation early warning accuracy in the prior art.
[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a landslide deformation early warning method, comprising the following steps:
[0006] S1, collect laser point cloud of the area to be monitored;
[0007] S2, separating the laser point cloud into an XY two-dimensional point cloud, an XZ two-dimensional point cloud and a YZ two-dimensional point cloud;
[0008] S3, extracting contours of the XY two-dimensional point cloud, the XZ two-dimensional point cloud, and the YZ two-dimensional point cloud respectively according to the reflection intensity to obtain a first contour structure, a second contour structure, and a third contour structure;
[0009] S4. Calculating a morphological value of the area to be monitored according to the first contour structure, the second contour structure and the third contour structure;
[0010] S5, extracting multiple frames of laser point clouds with morphological changes according to the morphological values of the area to be monitored corresponding to the laser point cloud;
[0011] S6, obtaining the deformation intensity of the landslide according to the first contour structure, the second contour structure and the third contour structure on each frame of the laser point cloud with morphological changes;
[0012] S7. When the landslide deformation intensity is greater than the intensity threshold, a landslide deformation warning is issued.
[0013] The beneficial effects of the present invention are as follows: the present invention separates the laser point cloud into an XY two-dimensional point cloud, an XZ two-dimensional point cloud and a YZ two-dimensional point cloud, realizes the analysis of the landslide deformation from three planes, extracts the contour structure of the three two-dimensional point clouds respectively according to the reflection intensity, simplifies the data volume, highlights the contour structure of the area to be monitored on the three planes, and then calculates the morphological value of the area to be monitored, finds out the multi-frame laser point cloud with morphological changes, obtains the landslide deformation intensity according to the three contour structures on each frame laser point cloud with morphological changes, and performs landslide deformation warning when the landslide deformation intensity is greater than the intensity threshold. The present invention realizes the comprehensive analysis of the landslide deformation intensity from three planes and improves the landslide deformation warning accuracy.
[0014] Furthermore, the extraction of contours from the XY two-dimensional point cloud, the XZ two-dimensional point cloud and the YZ two-dimensional point cloud in S3 includes the following sub-steps:
[0015] S31, performing clustering processing on each position point on the two-dimensional point cloud according to the distance of the reflection intensity of each position point on the two-dimensional point cloud, to obtain multiple clusters on the two-dimensional point cloud;
[0016] S32, calculating the average reflection intensity of each cluster on the two-dimensional point cloud;
[0017] S33, retaining the edge position points of each cluster on the two-dimensional point cloud to obtain the contour area of the cluster;
[0018] S34, marking the average reflection intensity on the contour area of the corresponding cluster to obtain the contour area of the marked cluster;
[0019] S35, forming a contour structure from the contour areas of each labeled cluster;
[0020] Among them, the contour is extracted from the XY two-dimensional point cloud to obtain a first contour structure, the contour is extracted from the XZ two-dimensional point cloud to obtain a second contour structure, and the contour is extracted from the YZ two-dimensional point cloud to obtain a third contour structure.
[0021] The beneficial effect of the above further scheme is: the present invention classifies the position points with similar reflection intensity in the two-dimensional point cloud into a cluster through clustering processing, retains only the edge position points, obtains the contour area of the cluster, extracts the average reflection intensity in a cluster, and reflects the reflection intensity level of the cluster.
[0022] Furthermore, the S4 comprises the following sub-steps:
[0023] S41, calculating a first contour morphology value according to contour areas of each labeled cluster on the first contour structure;
[0024] S42, calculating a second contour morphology value according to contour areas of each labeled cluster on the second contour structure;
[0025] S43, calculating a third contour morphology value according to the contour areas of each labeled cluster on the third contour structure;
[0026] S44. Calculate the morphology value of the area to be monitored according to the first contour morphology value, the second contour morphology value and the third contour morphology value.
[0027] Further, the formulas for calculating the first contour morphology value in S41, the second contour morphology value in S42, and the third contour morphology value in S43 are all:
[0028] , where γ is the first contour morphology value, the second contour morphology value or the third contour morphology value, (s i,j , r i,j ) is the coordinate of the jth position point on the contour area of the i-th labeled cluster on the first contour structure, the second contour structure or the third contour structure. In the first contour structure, s i,j is the x coordinate, r i,j is the y coordinate, in the second contour structure s i,j is the x coordinate, r i,j is the z coordinate, in the third contour structure s i,j is the y coordinate, r i,j is the z coordinate, re i is the average reflection intensity on the contour area of the ith labeled cluster, i and j are positive integers, N is the number of position points on the contour area of the ith labeled cluster on the first contour structure, the second contour structure or the third contour structure, and M is the number of contour areas of the labeled cluster on the first contour structure, the second contour structure or the third contour structure.
[0029] The beneficial effect of the above further solution is that the present invention adds each position point on the contour area of each labeled cluster , combined with the average reflection intensity of the cluster, the contour morphology value corresponding to the contour structure is obtained. The contour morphology value of the present invention includes the contour distribution characteristics and the average reflection intensity in the contour area of the labeled cluster, so that each contour structure has a specific contour morphology value.
[0030] Furthermore, the formula for calculating the morphological value of the area to be monitored in S44 is:
[0031] , where F is the morphological value of the area to be monitored, γ1 is the first contour morphological value, γ2 is the second contour morphological value, and γ3 is the third contour morphological value.
[0032] Furthermore, the S5 comprises the following sub-steps:
[0033] S51, calculating the distance between the laser point cloud at the current moment and the laser point cloud at the previous moment in terms of the morphological value of the area to be monitored: , where d t is the distance at the current moment, F t is the morphological value of the laser point cloud corresponding to the monitored area at the current moment, F t-1 is the morphological value of the laser point cloud corresponding to the monitored area at the previous moment, t is the current moment, and || is the absolute value operation;
[0034] S52: When the distances at multiple consecutive moments are greater than the distance threshold, the multi-frame laser point clouds corresponding to the multiple moments are multi-frame laser point clouds with morphological changes.
[0035] The beneficial effect of the above further scheme is that the morphological value of the monitored area of the present invention reflects the point cloud distribution of a frame of laser point cloud. Therefore, when the distance at multiple consecutive moments is greater than the distance threshold, it means that the laser point cloud has been changing during this period of time, thereby extracting multiple frames of laser point clouds with morphological changes.
[0036] Furthermore, the S6 comprises the following sub-steps:
[0037] S61, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the first contour structure to obtain a first landslide velocity;
[0038] S62, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the second contour structure to obtain a second landslide velocity;
[0039] S63, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the third contour structure to obtain a third landslide velocity;
[0040] S64. Calculate the landslide deformation intensity according to the first landslide velocity, the second landslide velocity and the third landslide velocity.
[0041] The beneficial effect of the above further scheme is that when calculating the landslide velocity, the present invention compares the contour structures of three aspects of two frames of laser point clouds at adjacent moments with morphological changes, thereby calculating the landslide velocity in each aspect, and comprehensively combining the landslide velocities in the three aspects to improve the accuracy of calculating the landslide deformation intensity.
[0042] Further, obtaining the first landslide speed in S61, obtaining the second landslide speed in S62, and obtaining the third landslide speed in S63 all include the following sub-steps:
[0043] A1. Calculate the area of the contour region of each labeled cluster on the contour structure, use the area as an element, and construct the contour structure vector;
[0044] A2. Calculate the difference of the contour structure vectors of two frames of laser point clouds at adjacent moments with morphological changes to obtain a difference vector;
[0045] A3. Calculate the landslide velocity based on the difference vector.
[0046] The beneficial effect of the above further scheme is: the present invention calculates the area of the contour area of each labeled cluster to form a contour structure vector, subtracts the contour structure vectors of two frames of laser point clouds at adjacent moments with morphological changes, and obtains a difference vector, which reflects the area change of the contour areas of the two, and calculates the landslide velocity through the difference vector.
[0047] Furthermore, the formula for calculating the landslide velocity in A3 is: , where v is the landslide velocity, z n is the nth difference in the difference vector, L is the number of differences in the difference vector, T is the time interval between two frames of laser point cloud at adjacent moments, n is a positive integer, and | | is an absolute value.
[0048] Furthermore, the formula for calculating the landslide deformation strength in S64 is:
[0049] , where g is the landslide deformation strength, v 1,k is the first landslide velocity at the kth moment, v 2,k is the second landslide velocity at the kth moment, v 3,k is the third landslide velocity at the kth moment, K is the statistical time length, θ is the proportional coefficient, and k is the moment number.
[0050] The beneficial effect of the above further solution is that the present invention comprehensively considers the three aspects of landslide velocity within the time length K to improve the calculation accuracy of landslide deformation intensity. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 The figure is a flow chart of a landslide deformation early warning method. DETAILED DESCRIPTION
[0052] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0053] like Figure 1 As shown, a landslide deformation early warning method comprises the following steps:
[0054] S1, collect laser point cloud of the area to be monitored;
[0055] S2, separating the laser point cloud into an XY two-dimensional point cloud, an XZ two-dimensional point cloud and a YZ two-dimensional point cloud;
[0056] S3, extracting contours of the XY two-dimensional point cloud, the XZ two-dimensional point cloud, and the YZ two-dimensional point cloud respectively according to the reflection intensity to obtain a first contour structure, a second contour structure, and a third contour structure;
[0057] S4. Calculating a morphological value of the area to be monitored according to the first contour structure, the second contour structure and the third contour structure;
[0058] S5, extracting multiple frames of laser point clouds with morphological changes according to the morphological values of the area to be monitored corresponding to the laser point cloud;
[0059] S6, obtaining the deformation intensity of the landslide according to the first contour structure, the second contour structure and the third contour structure on each frame of the laser point cloud with morphological changes;
[0060] S7. When the landslide deformation intensity is greater than the intensity threshold, a landslide deformation warning is issued.
[0061] In this embodiment, step S2 includes two implementation methods. The first one is to extract the corresponding two-dimensional coordinates of each point in the three-dimensional coordinates (X, Y, Z). The laser radar point cloud contains the three-dimensional coordinate information (X, Y, Z) and reflection intensity of the spatial point. When processing these data, the corresponding two-dimensional point cloud can be generated by selecting different dimension combinations. For example, to obtain an XY two-dimensional point cloud, it is only necessary to extract the X and Y coordinates of each point cloud; for an XZ two-dimensional point cloud, the X and Z coordinates are extracted; similarly, for a YZ two-dimensional point cloud, the Y and Z coordinates need to be extracted.
[0062] The second method is to project the 3D point cloud onto a 2D plane to obtain a 2D point cloud. For example, an XY 2D point cloud is obtained by projecting the 3D point cloud onto the XY 2D plane.
[0063] The extraction of contours from the XY two-dimensional point cloud, the XZ two-dimensional point cloud and the YZ two-dimensional point cloud in S3 includes the following sub-steps:
[0064] S31, performing clustering processing on each position point on the two-dimensional point cloud according to the distance of the reflection intensity of each position point on the two-dimensional point cloud, to obtain multiple clusters on the two-dimensional point cloud;
[0065] S32, calculating the average reflection intensity of each cluster on the two-dimensional point cloud;
[0066] S33, retaining the edge position points of each cluster on the two-dimensional point cloud to obtain the contour area of the cluster;
[0067] S34, marking the average reflection intensity on the contour area of the corresponding cluster to obtain the contour area of the marked cluster;
[0068] S35, forming a contour structure from the contour areas of each labeled cluster;
[0069] Among them, the contour is extracted from the XY two-dimensional point cloud to obtain a first contour structure, the contour is extracted from the XZ two-dimensional point cloud to obtain a second contour structure, and the contour is extracted from the YZ two-dimensional point cloud to obtain a third contour structure.
[0070] The present invention classifies the position points with similar reflection intensity in the two-dimensional point cloud into a cluster through clustering processing, retains only the edge position points, obtains the contour area of the cluster, extracts the average reflection intensity in a cluster, and reflects the reflection intensity level of the cluster.
[0071] The S4 comprises the following sub-steps:
[0072] S41, calculating a first contour morphology value according to contour areas of each labeled cluster on the first contour structure;
[0073] S42, calculating a second contour morphology value according to contour areas of each labeled cluster on the second contour structure;
[0074] S43, calculating a third contour morphology value according to the contour areas of each labeled cluster on the third contour structure;
[0075] S44. Calculate the morphology value of the area to be monitored according to the first contour morphology value, the second contour morphology value and the third contour morphology value.
[0076] The formulas for calculating the first contour morphology value in S41, the second contour morphology value in S42, and the third contour morphology value in S43 are all:
[0077] , where γ is the first contour morphology value, the second contour morphology value or the third contour morphology value, (s i,j , r i,j ) is the coordinate of the jth position point on the contour area of the i-th labeled cluster on the first contour structure, the second contour structure or the third contour structure. In the first contour structure, s i,j is the x coordinate, r i,j is the y coordinate, in the second contour structure s i,j is the x coordinate, r i,j is the z coordinate, in the third contour structure s i,j is the y coordinate, r i,j is the z coordinate, re iis the average reflection intensity on the contour area of the ith labeled cluster, i and j are positive integers, N is the number of position points on the contour area of the ith labeled cluster on the first contour structure, the second contour structure or the third contour structure, and M is the number of contour areas of the labeled cluster on the first contour structure, the second contour structure or the third contour structure.
[0078] The present invention adds each position point on the contour area of each labeled cluster , combined with the average reflection intensity of the cluster, the contour morphology value corresponding to the contour structure is obtained. The contour morphology value of the present invention includes the contour distribution characteristics and the average reflection intensity in the contour area of the labeled cluster, so that each contour structure has a specific contour morphology value.
[0079] The formula for calculating the morphological value of the area to be monitored in S44 is: , where F is the morphological value of the area to be monitored, γ1 is the first contour morphological value, γ2 is the second contour morphological value, and γ3 is the third contour morphological value.
[0080] The S5 comprises the following sub-steps:
[0081] S51, calculating the distance between the laser point cloud at the current moment and the laser point cloud at the previous moment in terms of the morphological value of the area to be monitored: , where d t is the distance at the current moment, F t is the morphological value of the laser point cloud corresponding to the monitored area at the current moment, F t-1 is the morphological value of the laser point cloud corresponding to the monitored area at the previous moment, t is the current moment, and || is the absolute value operation;
[0082] S52: When the distances at multiple consecutive moments are greater than the distance threshold, the multi-frame laser point clouds corresponding to the multiple moments are multi-frame laser point clouds with morphological changes.
[0083] The morphological value of the monitored area of the present invention reflects the point cloud distribution of a frame of laser point cloud. Therefore, when the distance at multiple consecutive moments is greater than the distance threshold, it means that the laser point cloud has been changing during this period of time, thereby extracting multiple frames of laser point cloud with morphological changes.
[0084] The S6 comprises the following sub-steps:
[0085] S61, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the first contour structure to obtain a first landslide velocity;
[0086] S62, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the second contour structure to obtain a second landslide velocity;
[0087] S63, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the third contour structure to obtain a third landslide velocity;
[0088] S64. Calculate the landslide deformation intensity according to the first landslide velocity, the second landslide velocity and the third landslide velocity.
[0089] When calculating the landslide velocity, the present invention compares the contour structures of three aspects of two frames of laser point clouds at adjacent moments with morphological changes, thereby calculating the landslide velocity in each aspect, and comprehensively analyzing the landslide velocities in the three aspects to improve the accuracy of calculating the landslide deformation intensity.
[0090] The steps of obtaining the first landslide speed in S61, obtaining the second landslide speed in S62, and obtaining the third landslide speed in S63 all include the following sub-steps:
[0091] A1. Calculate the area of the contour region of each labeled cluster on the contour structure, use the area as an element, and construct the contour structure vector;
[0092] A2. Calculate the difference of the contour structure vectors of two frames of laser point clouds at adjacent moments with morphological changes to obtain a difference vector;
[0093] In this embodiment, step A2 specifically includes: subtracting the contour structure vector of the laser point cloud at the current moment from the contour structure vector of the laser point cloud at the previous moment to obtain a difference vector;
[0094] A3. Calculate the landslide velocity based on the difference vector.
[0095] The present invention calculates the area of the contour region of each labeled cluster to form a contour structure vector, subtracts the contour structure vectors of two frames of laser point clouds at adjacent moments with morphological changes, and obtains a difference vector, which reflects the area change of the contour regions of the two. The landslide velocity is calculated through the difference vector.
[0096] The formula for calculating the landslide velocity in A3 is: , where v is the landslide velocity, z n is the nth difference in the difference vector, L is the number of differences in the difference vector, T is the time interval between two frames of laser point cloud at adjacent moments, n is a positive integer, and | | is an absolute value.
[0097] The formula for calculating the landslide deformation strength in S64 is:
[0098] , where g is the landslide deformation strength, v 1,k is the first landslide velocity at the kth moment, v 2,k is the second landslide velocity at the kth moment, v 3,kis the third landslide velocity at the kth moment, K is the statistical time length, θ is the proportional coefficient, and k is the moment number.
[0099] The invention comprehensively considers three aspects of landslide velocity within the time length K, thereby improving the calculation accuracy of landslide deformation strength.
[0100] In this embodiment, the intensity threshold and the distance threshold are specifically set based on experiments or experience.
[0101] The present invention separates the laser point cloud into an XY two-dimensional point cloud, an XZ two-dimensional point cloud and a YZ two-dimensional point cloud, realizes the analysis of the landslide deformation from three planes, extracts the contour structure of the three two-dimensional point clouds respectively according to the reflection intensity, simplifies the data volume, highlights the contour structure of the area to be monitored on the three planes, calculates the morphological value of the area to be monitored, finds out the multi-frame laser point cloud with morphological changes, obtains the landslide deformation intensity according to the three contour structures on each frame laser point cloud with morphological changes, and performs landslide deformation warning when the landslide deformation intensity is greater than the intensity threshold. The present invention realizes the comprehensive analysis of the landslide deformation intensity from three planes, and improves the landslide deformation warning accuracy.
[0102] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A landslide deformation early warning method, characterized in that: The following steps are involved: S1, collect laser point cloud of the area to be monitored; S2, separating the laser point cloud into an XY two-dimensional point cloud, an XZ two-dimensional point cloud and a YZ two-dimensional point cloud; S3, extracting contours of the XY two-dimensional point cloud, the XZ two-dimensional point cloud, and the YZ two-dimensional point cloud respectively according to the reflection intensity to obtain a first contour structure, a second contour structure, and a third contour structure; S4. Calculating a morphological value of the area to be monitored according to the first contour structure, the second contour structure and the third contour structure; S5, extracting multiple frames of laser point clouds with morphological changes according to the morphological values of the area to be monitored corresponding to the laser point cloud; S6, obtaining the deformation intensity of the landslide according to the first contour structure, the second contour structure and the third contour structure on each frame of the laser point cloud with morphological changes; S7. When the landslide deformation intensity is greater than the intensity threshold, a landslide deformation warning is issued; The extraction of contours from the XY two-dimensional point cloud, the XZ two-dimensional point cloud and the YZ two-dimensional point cloud in S3 includes the following sub-steps: S31, performing clustering processing on each position point on the two-dimensional point cloud according to the distance of the reflection intensity of each position point on the two-dimensional point cloud, to obtain multiple clusters on the two-dimensional point cloud; S32, calculating the average reflection intensity of each cluster on the two-dimensional point cloud; S33, retaining the edge position points of each cluster on the two-dimensional point cloud to obtain the contour area of the cluster; S34, marking the average reflection intensity on the contour area of the corresponding cluster to obtain the contour area of the marked cluster; S35, forming a contour structure from the contour areas of each labeled cluster; The first contour structure is obtained by extracting the contour of the XY two-dimensional point cloud, the second contour structure is obtained by extracting the contour of the XZ two-dimensional point cloud, and the third contour structure is obtained by extracting the contour of the YZ two-dimensional point cloud; The S4 comprises the following sub-steps: S41, calculating a first contour morphology value according to contour areas of each labeled cluster on the first contour structure; S42, calculating a second contour morphology value according to contour areas of each labeled cluster on the second contour structure; S43, calculating a third contour morphology value according to the contour areas of each labeled cluster on the third contour structure; S44, calculating a morphology value of the area to be monitored according to the first contour morphology value, the second contour morphology value and the third contour morphology value; The formulas for calculating the first contour morphology value in S41, the second contour morphology value in S42, and the third contour morphology value in S43 are all: , where γ is the first contour morphology value, the second contour morphology value or the third contour morphology value, (s i,j , r i,j ) is the coordinate of the jth position point on the contour area of the i-th labeled cluster on the first contour structure, the second contour structure or the third contour structure. In the first contour structure, s i,j is the x coordinate, r i,j is the y coordinate, in the second contour structure s i,j is the x coordinate, r i,j is the z coordinate, in the third contour structure s i,j is the y coordinate, r i,j is the z coordinate, re i is the average reflection intensity on the contour area of the i-th labeled cluster, i and j are positive integers, N is the number of position points on the contour area of the i-th labeled cluster on the first contour structure, the second contour structure or the third contour structure, and M is the number of contour areas of the labeled cluster on the first contour structure, the second contour structure or the third contour structure; The formula for calculating the morphological value of the area to be monitored in S44 is: , where F is the morphological value of the area to be monitored, γ1 is the first contour morphological value, γ2 is the second contour morphological value, and γ3 is the third contour morphological value.
2. The landslide deformation early warning method according to claim 1, characterized in that: The S5 comprises the following sub-steps: S51, calculating the distance between the laser point cloud at the current moment and the laser point cloud at the previous moment in terms of the morphological value of the area to be monitored: , where d t is the distance at the current moment, F t is the morphological value of the laser point cloud corresponding to the monitored area at the current moment, F t-1 is the morphological value of the laser point cloud corresponding to the monitored area at the previous moment, t is the current moment, and | | is the absolute value operation; S52: When the distances at multiple consecutive moments are greater than the distance threshold, the multi-frame laser point clouds corresponding to the multiple moments are multi-frame laser point clouds with morphological changes.
3. The landslide deformation early warning method according to claim 1, characterized in that: The S6 comprises the following sub-steps: S61, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the first contour structure to obtain a first landslide velocity; S62, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the second contour structure to obtain a second landslide velocity; S63, calculating the landslide velocity of two frames of laser point clouds at adjacent moments with morphological changes on the third contour structure to obtain a third landslide velocity; S64. Calculate the landslide deformation intensity according to the first landslide velocity, the second landslide velocity and the third landslide velocity.
4. The landslide deformation early warning method according to claim 3 is characterized in that: The steps of obtaining the first landslide speed in S61, obtaining the second landslide speed in S62, and obtaining the third landslide speed in S63 all include the following sub-steps: A1. Calculate the area of the contour region of each labeled cluster on the contour structure, use the area as an element, and construct the contour structure vector; A2. Calculate the difference of the contour structure vectors of two frames of laser point clouds at adjacent moments with morphological changes to obtain a difference vector; A3. Calculate the landslide velocity based on the difference vector.
5. The landslide deformation early warning method according to claim 4, characterized in that: The formula for calculating the landslide velocity in A3 is: , where v is the landslide velocity, z n is the nth difference in the difference vector, L is the number of differences in the difference vector, T is the time interval between two frames of laser point cloud at adjacent moments, n is a positive integer, and | | is an absolute value.
6. The landslide deformation early warning method according to claim 3 is characterized in that: The formula for calculating the landslide deformation strength in S64 is: , where g is the landslide deformation strength, v 1,k is the first landslide velocity at the kth moment, v 2,k is the second landslide velocity at the kth moment, v 3,k is the third landslide velocity at the kth moment, K is the statistical time length, θ is the proportional coefficient, and k is the moment number.
