A slope surface deformation rapid monitoring method based on InSAR technology and image recognition

Abstract

The present application belongs to the technical field of slope surface deformation monitoring, and discloses a kind of slope surface deformation rapid monitoring method based on InSAR technology and image recognition, including field detection and calculation analysis, as follows: S1.Field detection: S1.1.Lay out high-precision positioning reference station and image pickup reference point outside the slope area, then according to InSAR interpretation results, carry out slope deformation partition, analyze the main sliding direction and deformation boundary.The present application carries out slope stratum partition, deformation partition and potential slip surface determination, rock-soil mass parameter inversion and other work through surface wave and radar wave field detection, InSAR interpretation and rock-soil mass parameter database, intelligently searches slope instability mode and dynamically feedback stability state, the present scheme not only solves the technical idea of the lack or less of slope geological survey and field test, but also can truly and accurately reflect the field stability state, provides scientific basis and technical support for slope design and treatment measure making.

Description

Technical Field

[0001] This invention belongs to the field of slope surface deformation monitoring technology, specifically a rapid monitoring method for slope surface deformation based on InSAR technology and image recognition. Background Technology

[0002] Slope deformation and failure are the result of the combined effects of internal and external factors. Slope deformation and failure generally exhibit three stages: initial deformation, uniform deformation, and accelerated deformation. The macroscopic deformation characteristics are mainly determined by the microscopic adjustment of the slope's own stress. It is a comprehensive manifestation of the effects of the slope's own structure, groundwater, geological structure, human activities, and external environment (rainfall, earthquakes), and is a macroscopic representation of the slope's dynamic adjustment.

[0003] At present, slope surface deformation ( Figure 1 The main monitoring methods are conventional ones such as total stations and levels, as well as non-contact monitoring methods such as GPS high-precision positioning systems, InSAR remote sensing, micro-deformation monitoring systems, and 3D laser scanners. However, due to the inability of existing technologies to effectively coordinate airborne positioning, corresponding high-precision positioning analysis, and image interpretation, it is not possible to quickly detect surface centering changes and to understand and grasp the evolution and characteristics of slope deformation in a timely manner. Therefore, this application proposes a corresponding solution to address the above problems and provide scientific data for slope deformation zoning and spatiotemporal evolution analysis, as well as basic data for subsequent slope deformation prediction and forecasting. Summary of the Invention

[0004] The purpose of this invention is to provide a rapid monitoring method for slope surface deformation based on InSAR technology and image recognition, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a rapid monitoring method for slope surface deformation based on InSAR technology and image recognition, including on-site detection and calculation analysis, specifically comprising the following steps:

[0006] S1. On-site inspection:

[0007] S1.1. Set up high-precision positioning reference stations and image picking reference points outside the slope area, and then, based on the InSAR interpretation results, divide the slope deformation into zones and analyze the main sliding direction and deformation boundary.

[0008] S1.2. Based on the slope deformation zoning results, use drones to set up image pickup points within the slope area;

[0009] S1.3. The drone takes images of the slope area and stops at image pickup points from the top of the slope to the bottom of the slope according to the distribution of slope deformation zones.

[0010] S1.4. Each time the field detection, repeat S1.1-S1.4 process, and in the stay image pickup point, each time through 1 way observation line can;

[0011] S2. Calculation and analysis:

[0012] S2.1. Determine the SAR data source of the slope area, download the SAR image of the area before, and then use the two-track method to develop SAR image interpretation. After registration, de-noising, difference processing, the deformation rate of the permanent scatter point in the slope area and the deformation rate of the image pickup point are obtained;

[0013] S2.2. Based on the deformation rate of the permanent scatter point in the slope area, the deformation rate difference and the deformation rate statistical cluster analysis are applied to develop the early deformation zoning of the slope area;

[0014] S2.3. Based on the high-precision positioning reference station coordinates, the coordinates of the image pickup point measured by the unmanned aerial vehicle are applied to develop the positioning difference calculation of the deformation rate of each image pickup point;

[0015] S2.4. Using the image map of the slope area measured by the unmanned aerial vehicle oblique photography, the point cloud position of the whole slope is drawn, and the deformation rate of the point cloud is calculated;

[0016] S2.5. Combined with the deformation rate of the image pickup point of the slope before, the deformation rate of the image pickup point measured by the unmanned aerial vehicle, and other data, the least square optimization algorithm is applied to carry out the image coordinate and positioning coordinate conversion, and then the deformation rate of the image pickup point of the slope is unified;

[0017] S2.6. Using the unified deformation rate of the image pickup point of the slope and the point cloud position of the whole slope each time, the Lagrange interpolation method is applied to carry out the deformation rate of the point cloud of the whole slope, based on the SAR image interpretation results of the area before and the early deformation zoning, the deformation rate statistical cluster analysis of the point cloud of the whole slope is carried out, and the current deformation zoning of the slope area is carried out;

[0018] S2.6. Based on the current deformation zoning and deformation rate of the slope, the deformation space-time evolution model of the slope area is established.

[0019] Preferably, the image pickup point is arranged according to the main sliding direction and the deformation boundary result, taking into account the overall and local emphasis principle of the slope area, 2-3 observation lines parallel to the main sliding direction inside and outside the deformation boundary of the slope are arranged, and 3-4 image pickup points are arranged in each observation line.

[0020] Preferably, in S1.3, the unmanned aerial vehicle starts from the image pickup reference point, and after 1-2 minutes of static observation coordinates by using high-precision positioning equipment, it starts to ascend and hover over the slope area, and then stays for 1-2 minutes, and 5-10 images of the slope area are taken by using oblique photography;

[0021] After the UAV takes pictures of the slope area, according to the distribution of the slope deformation partition, the image pickup point is stopped for 1-2 minutes from the top of the slope to the bottom of the slope.

[0022] Preferably, in S2.3, based on the high-precision positioning reference station coordinates and the image pickup reference point, the point coordinates of the image pickup point of the slope monitored by the UAV are used to calculate the deformation of each image pickup point by using the positioning difference.

[0023] The beneficial effects of the present application are as follows:

[0024] 1. The present application carries out slope stratum partition, deformation partition, potential slip surface determination, rock and soil parameter inversion and other work through surface wave and radar wave field detection, InSAR interpretation and rock and soil parameter database, intelligently searches for slope instability mode and dynamically feeds back stability state. The present scheme not only solves the technical idea of the lack or less of slope geological survey and field test, but also truly and accurately reflects the field stability state, and provides a scientific basis and technical support for slope design and governance measure making.

[0025] 2. The present application establishes a slope deformation partition through the early InSAR remote sensing deformation recognition result; sets an image pickup point for the key part, uses airborne high-precision positioning and oblique photography equipment to carry out image recognition and interpretation; analyzes the space-time evolution law and change characteristics of the slope surface deformation through repeated image collection and interpretation, and further establishes an engineering slope deformation space-time evolution model. The present method not only solves the monitoring technical scheme before the establishment of the conventional monitoring of the slope, but also provides a scientific basis and technical support for the engineering slope deformation prediction and prediction and the later slope design and comprehensive treatment. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 It is a schematic diagram of the existing slope surface deformation monitoring method;

[0027] Figure 2 It is a schematic diagram of the slope deformation rapid monitoring method of the present application;

[0028] Figure 3 It is a schematic diagram of the slope deformation plane layout of the present application;

[0029] Figure 4 It is a schematic diagram of the slope deformation profile layout of the present application;

[0030] Figure 5 It is a schematic diagram of the slope deformation rate calculation method based on high-precision positioning of the present application;

[0031] Figure 6 It is a flow chart of the slope deformation rate calculation method based on InSAR and oblique photography of the present application;

[0032] Figure 7 The flow chart of the method for calculating the space-time evolution model of the slope deformation.

[0033] In the figure: 1, the boundary of the slope area; 2, the observation line; 3, the image pickup point of the slope area; 4, the image pickup reference point; 5, the high-precision positioning reference station; 6, the deformation boundary line; 7, the unmanned aerial vehicle; 8, the oblique camera; 9, the high-precision positioning device; 10, the deformation zone 4; 11, the deformation zone 3; 12, the deformation zone 2; 13, the deformation zone 1. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0035] As shown in the figure, the embodiment of the present application provides a slope surface deformation rapid monitoring method based on InSAR technology and image recognition, which detects the deformation of the slope surface in the field. Figures 1 to 7

[0036] 1) Outside the slope area, the high-precision positioning reference station and the image pickup reference point are arranged;

[0037] 2) According to the InSAR interpretation results, the slope deformation is zoned, the main sliding direction and the deformation boundary are analyzed;

[0038] 3) According to the slope deformation zoning results, the unmanned aerial vehicle is used to arrange the image pickup points in the slope area;

[0039] 4) According to the main sliding direction and the deformation boundary results, 2-3 observation lines parallel to the main sliding direction inside and outside the slope deformation boundary are arranged, and 3-4 image pickup points are arranged for each observation line, considering the overall and local principles of the slope area;

[0040] 5) After the unmanned aerial vehicle starts from the image pickup reference point and stays for 1-2 minutes to observe the coordinates statically by using the high-precision positioning device, the unmanned aerial vehicle starts to ascend and hovers over the slope area, stays for 1-2 minutes again, and then 5-10 images of the slope area are taken by using the oblique camera;

[0041] 6) After the unmanned aerial vehicle takes the images of the slope area, the image pickup points are stopped for 1-2 minutes in turn from the top of the slope to the bottom according to the distribution of the slope deformation zoning;

[0042] 7) Each time the field detection is repeated, the above 1) to 6) is repeated, and each time the image pickup point is stopped, it can pass through one observation line.

[0043] Calculation and analysis​

[0044] 1) Determine the SAR data source where the slope area is located, download the SAR image of the area before the area;

[0045] 2) Use the two-track method to carry out SAR image interpretation, obtain the deformation rate of the permanent scattering point in the slope area and the deformation rate of the image pickup point after registration, denoising, difference processing, etc.

[0046] 3) Based on the deformation rate of the permanent scattering point in the slope area, apply the deformation rate difference and deformation amount statistical cluster analysis to carry out the early deformation partition of the slope area;

[0047] 4) Based on the high-precision positioning reference station coordinates, apply the measured point coordinates of the slope image pickup point monitored by the unmanned aerial vehicle to carry out positioning difference calculation of the deformation rate of each image pickup point;

[0048] 5) Use the image map of the slope area measured by the unmanned aerial vehicle oblique photography to draw the point cloud position of the whole slope, and calculate the deformation rate of the point cloud;

[0049] 6) Combine the deformation rate of the slope image pickup point before, the deformation rate of the image pickup point measured by the unmanned aerial vehicle, etc. Apply the least squares optimization algorithm to carry out image coordinate and positioning coordinate conversion, and then unify the deformation rate of the slope image pickup point;

[0050] 7) Use the unified deformation rate of the slope image pickup point and the point cloud position of the whole slope each time to apply the Lagrange interpolation method to carry out the deformation rate of the whole slope point cloud;

[0051] 8) Based on the SAR image interpretation results and the early deformation partition of the area before, carry out the deformation rate statistical cluster analysis of the whole slope point cloud, and carry out the current deformation partition of the slope area;

[0052] 9) Based on the current deformation partition and deformation rate of the slope, establish the deformation space-time evolution model of the slope area.

[0053] It should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the term "include", "contain" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0054] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A rapid monitoring method for slope surface deformation based on InSAR technology and image recognition, characterized in that: Including on-site testing and computational analysis, the specific steps are as follows: S1. On-site inspection: S1.

1. Set up high-precision positioning reference stations and image picking reference points outside the slope area, and then, based on the InSAR interpretation results, divide the slope deformation into zones and analyze the main sliding direction and deformation boundary. S1.

2. Based on the slope deformation zoning results, use drones to set up image pickup points within the slope area; S1.

3. The drone takes images of the slope area and stops at image pickup points from the top of the slope to the bottom of the slope according to the distribution of slope deformation zones. S1.

4. For each on-site inspection, repeat the process from S1.1 to S1.4, and when stopping at the image acquisition point, pass through one observation line each time; S2. Calculation and Analysis: S2.

1. Determine the SAR data source for the slope area, download the previous SAR images of the area, and then use the two-track method to interpret the SAR images. After registration, de-noising, and differential processing, obtain the deformation rate of the permanent scattering points in the slope area and the deformation rate of the image picking points. S2.

2. Based on the deformation rate of permanent scattering points in the slope area, apply deformation rate difference and deformation statistical cluster analysis to carry out early deformation zoning of the slope area; S2.

3. Based on the coordinates of the high-precision positioning reference station, the coordinates of the measurement points of the slope image picking points monitored by the UAV are used to carry out positioning difference calculation of the deformation rate of each image picking point; S2.

4. Using the image map of the slope area measured by UAV oblique photography, draw the point cloud location of the entire slope area and calculate the deformation rate of the point cloud; S2.

5. Combining the deformation rate data of the previously acquired slope image points and the deformation rate data of the acquired image points measured by UAV, the least squares optimization algorithm is applied to transform the image coordinates and the positioning coordinates, thereby unifying the deformation rate of the slope image points. S2.

6. Using the deformation rate of the unified slope image pick-up points and the point cloud position of the entire slope area each time, the Lagrange interpolation method is applied to calculate the deformation rate of the point cloud of the entire slope area. Based on the previous SAR image interpretation results of the area and the previous deformation zoning, statistical cluster analysis of the deformation rate of the point cloud of the entire slope area is carried out to determine the current deformation zoning of the slope area. S2.

6. Based on the current deformation zoning and deformation rate results of the slope, establish a spatiotemporal evolution model of slope deformation.

2. The method for rapid monitoring of slope surface deformation based on InSAR technology and image recognition according to claim 1, characterized in that: The image pickup points are set up according to the main sliding direction and deformation boundary results, taking into account both the overall and local aspects of the slope area. There are 2-3 observation lines inside and outside the slope deformation boundary and parallel to the main sliding direction, with 3-4 image pickup points in each line.

3. The method for rapid monitoring of slope surface deformation based on InSAR technology and image recognition according to claim 1, characterized in that: In S1.3, the UAV starts up from the image acquisition reference point and uses a high-precision positioning device to stay for 1-2 minutes to statically observe the coordinates. Then, it takes off and hovers over the slope area, stays for 1-2 minutes again, and uses an oblique camera to take 5-10 images of the slope area. After the drone captures images of the slope area, it stops at each image pickup point for 1-2 minutes from the top to the bottom of the slope, according to the distribution of slope deformation zones.

4. The method for rapid monitoring of slope surface deformation based on InSAR technology and image recognition according to claim 1, characterized in that: In S2.3, based on the coordinates of the high-precision positioning reference station and the image picking reference point, the coordinates of the measurement points of the slope image picking points monitored by the UAV are applied, and the deformation of each image picking point is calculated using positioning difference.

Images