A safety evaluation method for pipeline excavation slopes based on three-dimensional reconstruction

Abstract

The present invention discloses a safety evaluation method for a pipeline excavation slope based on three-dimensional reconstruction, belonging to the field of slope landslide safety protection, and comprising the following steps: S1, arranging an image scanning system in the pipeline excavation slope area to collect slope images; S2, performing feature extraction and stereo matching on the collected slope images; S3, performing three-dimensional reconstruction on the landslide debris on the slope according to the extracted features and stereo matching results; S4, identifying the types, sizes, quantities, and position features of the landslide debris according to the three-dimensional reconstruction results; S5, analyzing and evaluating the safety of the pipeline slope excavation according to the feature information identified in step S4, which can realize real-time monitoring of slope landslides, can timely detect changes in the slope, provide timely and accurate information for early warning and response, automatically judge the stability state of the slope, and issue an early warning signal in time when abnormalities are found.

Description

Technical Field

[0001] The present invention relates to the field of slope landslide safety protection, and particularly to a safety evaluation method for pipeline excavation slopes based on three-dimensional reconstruction. Background Art

[0002] Pipeline projects often pass through different geological environments, including soft soil layers, rock layers, and mixed soil layers. The changes in these geological conditions can lead to significant differences in slope stability. Especially in soft soil and weathered rock layers, the landslide risk is higher. Natural factors such as rainfall, earthquake, weathering, and groundwater changes will have a significant impact on slope stability. Moreover, the pipeline itself will also cause disturbances to the slope during the excavation construction process, resulting in landslides. Any damage caused by landslides will bring serious impacts, including economic losses, environmental pollution, and threats to personal safety. Therefore, ensuring slope stability is crucial. Currently, there is no relevant technology for safety evaluation of pipeline excavation slopes. Summary of the Invention

[0003] In view of the above problems, the present invention provides a safety evaluation method for pipeline excavation slopes based on three-dimensional reconstruction. By using three-dimensional reconstruction technology, real-time monitoring of slopes can be achieved, changes in slopes can be detected in a timely manner, and timely and accurate information can be provided for early warning and response to ensure the safety of slope operations.

[0004] The technical solution of the present invention is as follows:

[0005] A safety evaluation method for pipeline excavation slopes based on three-dimensional reconstruction, comprising the following steps:

[0006] S1. Deploy an image scanning system in the pipeline excavation slope area to collect slope images;

[0007] S2. Perform feature extraction and stereo matching on the collected slope images;

[0008] S3. Perform three-dimensional reconstruction on the landslide debris on the slope according to the extracted features and stereo matching results;

[0009] S4. Identify the types, sizes, quantities, and location characteristics of the landslide debris according to the three-dimensional reconstruction results;

[0010] S5. Analyze and evaluate the safety of pipeline slope excavation according to the feature information identified in step S4.

[0011] In step S1, the image scanning system is a plurality of laser scanners that fully cover the pipeline excavation slope area.

[0012] In step S1, the method of hand-eye calibration is used for precise positioning and recognition when collecting slope images. The specific method is as follows:

[0013] 1) Define an arbitrary world coordinate system on the target to be measured. The conversion formula from the world coordinate system to the scanning system coordinate system is as follows:

[0014]

[0015] where (Wx, Wy, Wz, 1) T is the homogeneous coordinate of the target point in the world coordinate system; (Cx, Cy, Cz, 1) T is the homogeneous coordinate of the target point in the camera coordinate system; E is a 3x3 orthogonal unit matrix; S is a 3x1 translation vector; 0 is a 3x1 zero vector;

[0016] 3) The conversion formula from the scanning system coordinate system to the imaging plane coordinate system is as follows:

[0017]

[0018] where (X, Y) T is the homogeneous coordinate in the imaging plane coordinate system, and f is the focal length.

[0019] In step S2, the stereo matching formula is as follows:

[0020]

[0021] Census(x,y,z) = ming(Census l (x,y), Census r (x - d,y))

[0022]

[0023] C(x,y,z) = αmin(T c , Census(x,y,z)) + (1 - α)min(T g , g(x,y,z))

[0024] S(p,z) = ∑ r L r (p,z)

[0025] In the formula: I(p), I(q) are the gray values of the image, I(p) is the average of the pixel gray values within the window, ξ is the threshold for gray binarization, Census(x, y, z) is the color difference between the corresponding pixel points in the left and right images, G(x, y, z) is the matching cost of the point (x, y) in the image, at this time the disparity is z, IL(x, y) is the gray value of the pixel point in the left image, IR(x, y, z) is the gray value of the corresponding homologous pixel point in the right image, α is the weight factor to balance the proportion of the Census algorithm and the gradient, Tc, Tg are the truncation values, Lr (p, z) is the path cost of a certain path.

[0026] In step S5, according to the feature recognition result, determine the threat level of the pipeline excavation slope to personnel. After discovering potential threats, send out early warning information and intervene in the construction personnel.

[0027] The beneficial effects of the present invention are as follows:

[0028] Capture images through a laser scanner, without direct contact with the object to be measured, avoiding physical damage or influence that may be caused by traditional contact measurement; by calculating the parallax information of pixel points in the image, the three-dimensional information of the object to be measured, including shape, size, and position, etc., can be accurately obtained, thereby realizing high-precision monitoring; realizing real-time monitoring of slope landslides, by continuously shooting and analyzing images, the changes of the slope can be timely detected, providing timely and accurate information for early warning and response, automatically judging the stability state of the slope, and sending out early warning signals in time when abnormalities are found. Description of the Drawings

[0029] Figure 1 It is a flowchart of a pipeline excavation slope safety evaluation method based on three-dimensional reconstruction according to an embodiment of the present invention. Detailed Embodiments

[0030] The following further describes the embodiments of the present invention with reference to the drawings.

[0031] Embodiment:

[0032] As Figure 1 shown, a pipeline excavation slope safety evaluation method based on three-dimensional reconstruction includes the following steps:

[0033] S1. Install an image scanning system in the pipeline excavation slope area to collect slope images;

[0034] Specifically, install a laser scanning system. Install a laser scanning system in the key areas of the pipeline excavation slope, fix and install the laser scanner with a movable bracket, adjust the position and angle of the equipment so that they can clearly capture the changes of the slope, ensure that the system can cover the areas where landslides and falling objects may occur. If full coverage cannot be achieved, add this system additionally to ensure full coverage of the working area. First, perform hand-eye calibration on the laser scanner to ensure that the data obtained by it can be accurately mapped to the coordinate system of the early warning system equipment, thereby realizing accurate target positioning and recognition; secondly, use the laser scanner to scan the slope to obtain image information.

[0035] The specific method of hand-eye calibration is as follows:

[0036] 1) Define an arbitrary world coordinate system on the target to be measured, and convert it from the world coordinate system to the scanning system coordinate system. The conversion formula is as follows:

[0037]

[0038] where (Wx, Wy, Wz, 1) T is the homogeneous coordinate of the target point in the world coordinate system; (Cx, Cy, Cz, 1) T is the homogeneous coordinate of the target point in the camera coordinate system; E is a 3x3 orthogonal unit matrix; S is a 3x1 translation vector; 0 is a 3x1 zero vector;

[0039] 4) Convert from the scanning system coordinate system to the imaging plane coordinate system. The conversion formula is as follows:

[0040]

[0041] where (X, Y) T is the homogeneous coordinate in the imaging plane coordinate system, and f is the focal length.

[0042] Due to the camera accuracy, the origin of the imaging plane coordinate system generally deviates from the origin of the image coordinate system. If the coordinates of the origin of the imaging plane coordinate system in the image coordinate system are known as (u0, v0), then this conversion relationship can be expressed as:

[0043]

[0044] where M1 is the camera internal parameter matrix; M2 is the camera external parameter matrix, which is determined by the orientation of the camera relative to the world coordinate system.

[0045] S2. Perform feature extraction and stereo matching on the collected slope images;

[0046] During the matching, the Census algorithm is used for processing. The formula of the Census algorithm is as follows:

[0047]

[0048] where I(p) and I(q) are the gray values of the image, I(p) is the average value of the pixel gray values in the window, and ξ is the threshold for gray binaryzation.

[0049] The corresponding pixel points in the left and right images have a similar relationship in gray value magnitude. Therefore, the matching degree between them can be determined by comparing the Census transform codes of the pixel points in the left and right images:

[0050] Census(x,y,z) = ming(Census l (x,y),Censusr (x - d, y))

[0051] Among them, Census(x, y, z) is the color difference between corresponding pixel points in the left and right images.

[0052] By using the gradient transformation operator, the similarity between images can be calculated more accurately, and its formula is as follows:

[0053]

[0054] Among them, G(x, y, z) is the matching cost of (x, y) in the image. At this time, the disparity value is z, IL(x, y) is the gray value of this pixel point in the left image, and IR(x, y, z) is the gray value of the corresponding homologous pixel point of this pixel point in the right image.

[0055] The Census transformation algorithm has excellent robustness and can effectively resist the gray - level changes caused by uneven illumination. The gradient transformation can better preserve the details of the image edges. Combining the two and using the matching cost for normalization processing, the formula is expressed as follows:

[0056] C(x, y, z) = αmin(T c , Census(x, y, z))+(1 - α)min(T g , g(x, y, z))

[0057] Among them, α is the weight factor, balancing the proportion of the Census algorithm and the gradient, and Tc and Tg are truncation values.

[0058] Aggregate the matching costs of each pixel point at different disparities to obtain the path value of each path. Then, accumulate the cost values of all possible matching points of this pixel point in the left and right images to calculate the total matching cost. Total path cost:

[0059] S(p, z) = ∑ r L r (p, z)

[0060] Among them, L r (p, z) is the path cost of a certain path.

[0061] S3. For the extracted features and stereo matching results, perform 3D reconstruction on the landslide debris on the slope;

[0062] S4. According to the 3D reconstruction results, identify the types, sizes, quantities, and location features of the landslide debris;

[0063] S5. According to the feature information identified in step S4, analyze and evaluate the safety of the pipeline slope excavation.

[0064] Based on the recognition results, determine whether the excavation slope of the pipeline poses a threat to personnel, determine whether landslide debris poses a threat to the excavation slope of the pipeline, determine the safe area at the slope. If workers enter the non-safe area, an early warning signal is issued, and the construction personnel quickly leave the area. If potential threats are discovered, an early warning signal is issued in a timely manner to remind the construction personnel to take corresponding measures.

[0065] The above embodiments only represent the specific implementation manners of the present invention, and the description thereof is relatively specific and detailed. However, it should not be construed as a limitation to the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. A safety evaluation method for the slope of a pipeline excavation based on three-dimensional reconstruction, characterized in that, It includes the following steps: S1. Deploy an image scanning system in the slope area of the pipeline excavation to collect slope images; S2. Extract features and perform stereo matching on the collected slope images; S3. Perform 3D reconstruction on the landslide debris on the slope according to the extracted features and stereo matching results; S4. Identify the types, sizes, quantities, and location features of the landslide debris according to the 3D reconstruction results; S5. Analyze and evaluate the safety of the pipeline slope excavation according to the feature information identified in step S4; In step S1, the method of hand-eye calibration is used for precise positioning and identification when collecting slope images. The specific method is as follows: 1) Define an arbitrary world coordinate system on the measured target, and convert it from the world coordinate system to the scanning system coordinate system. The conversion formula is as follows: where (Wx, Wy, Wz, 1) T is the homogeneous coordinate of the target point in the world coordinate system; (Cx, Cy, Cz, 1) T is the homogeneous coordinate of the target point in the camera coordinate system; E is a 3x3 orthogonal unit matrix; S is a 3x1 translation vector; 0 is a 3x1 zero vector; 2) Convert from the scanning system coordinate system to the imaging plane coordinate system. The conversion formula is as follows: where (X, Y) T are homogeneous coordinates in the imaging plane coordinate system, and f is the focal length; In step S2, the stereo matching formula is as follows: Census(x,y,z) = min(Census l (x,y), Census r (x - d,y)) C(x, y, z) = α min(T c , Census(x, y, z)) + (1 - α) min(T g , g(x, y, z)) S(p,z) = ∑ r L r (p,z) Where: I(p) and I(q) are the gray values of the image, I(p) is the average of the pixel gray values within the window, ξ is the threshold for gray binarization, Census(x, y, z) is the color difference between corresponding pixel points in the left and right images, G(x, y, z) is the matching cost of the point (x, y) in the image, at this time the disparity is z, IL(x, y) is the gray value of this pixel point in the left image, IR(x, y, z) is the gray value of the corresponding homologous pixel point of this pixel point in the right image, α is the weight factor to balance the proportion of the Census algorithm and the gradient, Tc and Tg are the truncation values, L r (p, z) is the path cost of a certain path.

2. The method for safety evaluation of a pipeline excavation slope based on three-dimensional reconstruction according to claim 1, wherein, In step S1, the image scanning system is multiple laser scanners that fully cover the pipeline excavation slope area.

3. The method for safety evaluation of a pipeline excavation slope based on three-dimensional reconstruction according to claim 1, characterized in that In step S5, according to the feature recognition results, judge the threat level of the pipeline excavation slope to personnel. After discovering potential threats, issue warning information to intervene in the construction personnel.

Images