SFM-based river model test topographic survey method
Through the SFM method combined with photogrammetry and computer vision technology, the rapid, efficient and accurate reconstruction of the three-dimensional terrain of the riverbed is achieved, solving the problems of low efficiency and insufficient accuracy of traditional methods, and the silt changes of the riverbed can be accurately obtained.
Patent Information
- Application Number
- CN202410936877.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-12
AI Technical Summary
Traditional riverbed three-dimensional topography measurement methods cannot effectively reflect the topography data of the entire research area, and there are problems of large workload and low efficiency. There are unsolved scientific problems in the combination of image three-dimensional reconstruction technology with riverbed terrain in river engineering model experiments.
Using SFM-based method, the feature points are extracted and matched by acquiring image sequences, sparse point cloud reconstruction is carried out, and dense point cloud reconstruction is carried out using CMVS and PMVS methods. Combining photogrammetry and computer vision technology, the precise remediation of three-dimensional terrain is achieved. and grid interpolation of the reconstructed point cloud data to obtain relevant parameters such as riverbed elevation.
It realizes fast, efficient and accurate three-dimensional terrain reconstruction, which can accurately obtain the silt changes of the riverbed. It is simple and convenient compared to traditional methods, and the measurement accuracy is close to that of the lidar.
Smart Images

Figure CN118941737A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of riverbed three-dimensional topography measurement, and in particular to a river engineering model test topography measurement method based on SFM. Background Art
[0002] The evolution of riverbed scouring and silting caused by water and sand movement is a common phenomenon in nature, which often causes practical engineering problems such as river channel silting, river bank deformation, coastal retreat, and reduced reservoir silting capacity. It is closely related to the design, construction and operation of water conservancy projects. Researching accurate and efficient riverbed three-dimensional topographic measurement methods and analyzing the changes in scouring and silting volume are of great significance to river engineering model tests and practical engineering applications.
[0003] Traditional riverbed 3D topographic measurement methods mainly rely on tools such as probes, steel rulers, levels, theodolites, and total stations. However, these methods are single-point measurements and cannot reflect the topographic data of the entire study area. They have limitations such as large workload and low efficiency.
[0004] Nowadays, the practicality of new topographic measurement methods such as photoelectric reflection topographs, resistive topographs, tracking topographs, ultrasonic topographs, laser scanners, and close-range photogrammetry is constantly improving. With the rapid development of photogrammetry and computer vision technology, the three-dimensional topographic reconstruction method through camera technology has made it possible to study the topography of river engineering model tests. In the past, in the study of gully erosion, through a comprehensive comparison of different measurement methods such as SFM, total station, and laser profiler, it was pointed out that the SFM measurement method has advantages in cost and time. The effectiveness of SFM has been proven in land surveying, and many research results have also confirmed its practicality in the laboratory. Its measurement accuracy is similar to that of other topographic measurement technologies such as lidar. However, there are unresolved scientific issues on how to combine image three-dimensional reconstruction technology with riverbed topography in river engineering model tests. Summary of the invention
[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a river engineering model test topography measurement method based on SFM, which combines the three-dimensional terrain reconstruction based on SFM (Structure From Motion) with data analysis, and applies it to the study of scour terrain evolution. The three-dimensional terrain point cloud is grid interpolated, which can accurately and efficiently obtain relevant parameters such as riverbed elevation to complete terrain measurement.
[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:
[0007] A topographic measurement method for river engineering model test based on SFM, the key of which is to include the following steps:
[0008] Step 1, obtaining an image sequence of the river engineering model test terrain to be tested;
[0009] Step 2: Use the SFM method to extract and match the feature points of the acquired image sequence, and reconstruct the sparse point cloud of the image sequence. Then use the CMVS and PMVS methods to reconstruct the dense point cloud of the sparse reconstruction results to achieve three-dimensional reconstruction of the terrain of the river engineering model test to be tested.
[0010] Step 3: Remove the point cloud data outside the research area from the reconstructed 3D point cloud data, and perform fine grid division on the research area;
[0011] Step 4: insert control points to obtain the actual three-dimensional coordinates of the terrain of the river engineering model test terrain to be tested, and interpolate the three-dimensional coordinates to the bed grid to obtain a three-dimensional river engineering model that meets the boundary constraints.
[0012] Furthermore, when acquiring the image sequence in step 1, a plurality of calibration points are respectively set on both sides of the river engineering model test terrain to be measured.
[0013] Furthermore, the calibration points are black and white checkerboard squares.
[0014] Furthermore, in the image sequence acquired in step 1, the overlap of images is not less than 80%, and the total number of images is not less than 200.
[0015] Furthermore, in step 1, a smart phone or a digital camera is used to acquire an image sequence.
[0016] Furthermore, in step 2, the SFM method is used to extract and match feature points of the acquired image sequence to achieve the three-dimensional reconstruction of the terrain of the river engineering model test to be tested. The specific steps are as follows:
[0017] Step 2.1, input the acquired image sequence, and use SIFT operator and SURF operator to detect and extract feature points of each image in the image sequence;
[0018] Step 2.2, using the KD tree proximity search algorithm, the feature points extracted from different images are matched to obtain feature points with the same name;
[0019] Step 2.3, according to the two-dimensional coordinates of the camera calibration points and the related geometric constraint relationship equations, calculate the external camera coordinates, direction angles and three-dimensional spatial coordinates;
[0020] Step 2.4, reconstruct the image sparse point cloud, and expand the coefficient point cloud through bundle adjustment;
[0021] Step 2.5: Cluster the sparsely reconstructed point cloud data using the CMVS method, and then use the PMVS method to generate a dense point cloud through matching, diffusion, and filtering under the constraints of local photometric consistency and global visibility, complete the dense point cloud reconstruction, and realize the three-dimensional reconstruction of the river engineering model test terrain to be tested.
[0022] Furthermore, in step 2.2, before performing feature point matching, the number of feature points of each image is compared according to a preset threshold, images with feature point numbers greater than the preset threshold are subjected to feature point matching, and images with feature point numbers less than the preset threshold are deleted.
[0023] Furthermore, the calculation formula for calculating the external camera coordinates in step 2.3 is:
[0024] mq=A[RT]P,
[0025] in, is the camera’s intrinsic parameter matrix, m is the scale correction factor, q=(u,v,1) T are the coordinates in the image coordinate system, (u0, v0) are the coordinates of the projection center in the image plane, (α, β) are the scale parameters in the u and v directions respectively, c is the pixel distortion parameter, P is the terrain coordinate in the world coordinate system, and R and T are the external parameters of the camera.
[0026] Furthermore, the interpolation result expression of the three-dimensional coordinates interpolated onto the bed surface grid in step 4 is as follows:
[0027]
[0028] Among them, (x0, y0, z0) represents the point to be interpolated, (x i ,y i ,z i ) is the i-th point closest to the point to be interpolated (x0, y0, z0), is the weight, is the distance between two points, α is the adjustment parameter, i=1,2,…,n, and n is the number of points closest to the interpolation point (x0, y0, z0).
[0029] Furthermore, the boundary constraints include:
[0030] Continuity constraint: The interpolation result has no discontinuous jumps at the boundary and in the entire area;
[0031] Smoothness constraint: The curve or surface of the interpolation result is smooth, without sharp corners or drastic fluctuations.
[0032] The remarkable effects of the present invention are:
[0033] The present invention combines photogrammetry with computer vision technology to quickly, efficiently and accurately reconstruct three-dimensional terrain;
[0034] From the large amount of dense point cloud data carried in the acquired photos, the point cloud data of non-research areas are screened and eliminated to ensure the accuracy of its three-dimensional reconstruction;
[0035] The study area is divided into fine grids, and the physical scale of the grid is set to 1mm×1mm. On this basis, the three-dimensional coordinates of the actual terrain are obtained by inserting control points, and the three-dimensional coordinates are interpolated on the bed grid to obtain the three-dimensional riverbed topography. This allows the changes before and after scouring and silting to be accurately obtained, which is simple and convenient compared to single-point measurement with a laser rangefinder. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 is a flow chart of the method of the present invention;
[0037] Figure 2 This is the principle flow chart of SFM 3D reconstruction;
[0038] Figure 3 It is a schematic diagram of the process of 3D reconstruction using the VisualSFM platform;
[0039] Figure 4 This is a schematic diagram of the VisualSFM platform when inputting an image;
[0040] Figure 5 This is a diagram of the feature point extraction and matching process of the VisualSFM platform;
[0041] Figure 6 It is the sparse reconstruction map of the VisualSFM platform;
[0042] Figure 7 It is a dense reconstruction map of the VisualSFM platform;
[0043] Figure 8 is a dense point cloud data graph in this embodiment;
[0044] Fig. 9 is the point cloud reconstruction image of this embodiment;
[0045] Fig.10 is a grid interpolation graph of this embodiment;
[0046] Fig.11 It is a comparison chart of the calculated value of the method of the present invention and the laser measured value. DETAILED DESCRIPTION
[0047] The specific implementation manner and working principle of the present invention are further described in detail below with reference to the accompanying drawings.
[0048] like Figure 1 As shown, this embodiment proposes a river engineering model test topographic measurement method based on SFM, and the specific steps are as follows:
[0049] Step 1: Use a smart phone or a digital camera to obtain an image sequence of the test terrain of the river engineering model to be tested;
[0050] When acquiring the image sequence, multiple calibration points are set on both sides of the river engineering model test terrain to be tested. The calibration points are black and white checkerboard squares. The black and white checkerboard arrangement is used for control, on the one hand, it can enhance the recognition of the image acquisition device, and on the other hand, it can also improve the accuracy of the evaluation and reconstruction point cloud.
[0051] In the acquired image sequence, the image overlap is not less than 80%, and the total number of images is not less than 200.
[0052] During the image acquisition process, the higher the image resolution, the more image information can be captured, and the accuracy of the shooting measurement can be improved by maximizing the depth of field. The orientation angle and coordinates of each photo in the process of reconstructing the 3D model include the spatial posture parameters of the camera corresponding to each photo.
[0053] Step 2: Use the SFM method to extract and match the feature points of the acquired image sequence, and reconstruct the sparse point cloud of the image sequence. Then use the CMVS and PMVS methods to reconstruct the dense point cloud of the sparse reconstruction results to achieve three-dimensional reconstruction of the terrain of the river engineering model test to be tested.
[0054] The SFM method can fully automatically reconstruct a 3D scene from a 2D image and simultaneously retrieve the corresponding camera geometry in an arbitrary coordinate system.
[0055] The typical workflow of the SFM method includes the following steps: first, identification and matching of homologous image points in overlapping photos; second, reconstruction of the geometric image acquisition configuration and the three-dimensional coordinates corresponding to the matching image points (sparse point cloud) using iterative BA; and third, dense matching of the sparse point cloud based on the reconstructed image geometric network.
[0056] The SFM method essentially relies on automated processing tools that can be provided by different non-commercial or commercial software packages. Its advantage is that it can be used at almost any scale. The core of the method is to identify and extract the feature points of each photo, and match the feature points on different photos to obtain the same-name points. Then, based on the two-dimensional coordinates of the camera calibration points and the related geometric constraint relationship equations, the external camera coordinates, direction angles and spatial three-dimensional coordinates are calculated for three-dimensional reconstruction of the terrain.
[0057] See attached Figure 2The SFM method extracts and matches feature points of the acquired image sequence to achieve the three-dimensional reconstruction of the terrain of the river engineering model test to be tested. The specific steps are as follows:
[0058] Step 2.1, input the acquired image sequence, and use the SIFT operator and SURF operator of the SIFT algorithm to detect and extract feature points of each image in the image sequence;
[0059] Among them, the SIFT (Scale-invariant feature transform) key point detection method identifies and extracts feature points. The advantage of this algorithm is that it will not be unstable due to different shooting conditions.
[0060] Step 2.2, using the KD tree proximity search algorithm, so that the known point can quickly and effectively find the closest d-dimensional space point for feature point matching, so as to match the feature points extracted from different images to obtain feature points with the same name;
[0061] During the matching process, when the number of matching points between images reaches above the preset threshold, they will be included in the SFM main program calculation; otherwise, if it is lower than the preset threshold, the image will be discarded.
[0062] Step 2.3, according to the two-dimensional coordinates of the camera calibration points and the related geometric constraint relationship equations, calculate the external camera coordinates, direction angles and three-dimensional spatial coordinates;
[0063] The world coordinate system is a three-dimensional coordinate system representing the real world. Let the terrain coordinates in the world coordinate system be P W =(X W ,Y W ,Z W ,1) T .
[0064] The image coordinate system is the two-dimensional coordinate system that transforms the three-dimensional coordinates presented by the camera into two-dimensional coordinates after being projected onto the screen. Let the coordinates in the image coordinate system be q = (u, v, 1) T .
[0065] The basic formula for camera coordinate transformation is: mq = A[RT]P,
[0066] in, is the camera’s intrinsic parameter matrix, m is the scale correction factor, q=(u,v,1) Tis the coordinate in the image coordinate system, (u0, v0) is the coordinate of the projection center in the image plane, (α, β) are the scale parameters in the u and v directions respectively, c is the pixel distortion parameter, P is the terrain coordinate in the world coordinate system, R and T are the external parameters of the camera, R represents the rotation change in the two coordinate systems, including 3 independent parameters. T represents the translation change in the two coordinate systems, including 3 independent parameters.
[0067] The camera lens may have errors and distortions due to production and processing. Considering that the camera distortion has radial distortion and tangential distortion, it is necessary to add two more image deformation parameters K1, K2, K3, and K4.
[0068] From the above, we can see that there are a total of 16 parameters, including 5 intrinsic parameters, 6 extrinsic parameters, 4 distortion parameters, and 1 scale parameter.
[0069] For the convenience of description, the 16 parameters are combined into a vector parameter to define the projection operator F:
[0070] q=F(θ,P,J,G)
[0071] Therefore, let P be a point in the world coordinate system, θ represents the camera, and the camera takes each photo to obtain the image coordinates q1…q n . . Where J is a parameter vector consisting of intrinsic parameters, distortion parameters and scale parameters, and G is a parameter vector consisting of extrinsic parameters. Each parameter can be solved by the following formula:
[0072]
[0073] Assume that the fixed point P1 in the world coordinate system appears n times in n pictures. Based on the equality of coordinates, n independent equations can be established. When there are enough matching points in multiple pictures, the camera parameters can be solved, the optimal projection relationship can be determined, and the world coordinate system can be inversely calculated.
[0074] Assume that in the camera θ shooting, there are b matching points Ci (1≤i≤b) in the a picture. The matching point is projected on the camera as Qi (1≤i≤b), and the matching point is projected as Yi (1≤i≤b). When the error between Qi and Yi is very small, Ci can be considered as the world coordinate system of the matching point.
[0075] Step 2.4, reconstruct the image sparse point cloud, and expand the coefficient point cloud through bundle adjustment;
[0076] Step 2.5: Cluster the sparsely reconstructed point cloud data using the CMVS method, and then use the PMVS method to generate a dense point cloud through matching, diffusion, and filtering under the constraints of local photometric consistency and global visibility, complete the dense point cloud reconstruction, and realize the three-dimensional reconstruction of the river engineering model test terrain to be tested.
[0077] In this embodiment, the VisualSFM platform is selected to implement the above-mentioned 3D reconstruction process. VisualSFM is a GUI application that uses motion structure (SFM) for 3D reconstruction. It runs quickly by utilizing multi-core parallelism for feature detection, feature matching, and package adjustment. It can realize the entire SFM calculation process from importing image coordinates to calibrating and obtaining point cloud data, thereby improving the calculation speed. The platform also integrates PMVS and CMVS algorithms.
[0078] The operation process of the VisualSFM platform is as follows Figure 3 As shown, the details are as follows:
[0079] (1) Image input - Import image sequence Click [1] and select multiple images. Figure 4 shown.
[0080] (2) Feature point extraction and matching - Click [2] to extract feature points for each image. At the same time, feature point and feature point matching files will be generated in the image folder. Therefore, for the same image, feature point extraction only needs to be done once, which greatly saves computing time. Figure 5 shown.
[0081] (3) Sparse reconstruction - Click [3] to start sparse reconstruction. The purpose is to restore the 3D coordinates of the matched feature points and the camera’s posture information. Figure 6 shown.
[0082] (4) Dense reconstruction - Click [4] to start the dense point cloud reconstruction through PMVS and CMVS algorithms to achieve the 3D reconstruction of the terrain of the river engineering model test to be tested. Then the save file window pops up, select the save path and file name, and complete the saving of the reconstructed 3D point cloud data. Figure 7 shown.
[0083] Step 3: Remove the point cloud data outside the research area from the reconstructed 3D point cloud data, and perform fine grid division on the research area;
[0084] Step 4: insert control points to obtain the actual three-dimensional coordinates of the terrain of the river engineering model test terrain to be tested, and interpolate the three-dimensional coordinates to the bed grid to obtain a three-dimensional river engineering model that meets the boundary constraints.
[0085] The boundary constraints include:
[0086] Continuity constraint: The interpolation result has no discontinuous jumps at the boundary and in the entire area;
[0087] Smoothness constraint: The curve or surface of the interpolation result is smooth, without sharp corners or drastic fluctuations.
[0088] The interpolation principle is to find the data point closest to the point to be interpolated among the known data points, assign the value of the data point to the point to be interpolated, interpolate the irregular data, and generate the interpolation on a regular grid.
[0089] Different weights are assigned to multiple neighboring points within a certain range for interpolation, the original data is smoothed, and noise or outliers are removed to reduce drastic changes in the data.
[0090] The interpolation result expression is as follows:
[0091]
[0092] Among them, (x0, y0, z0) represents the point to be interpolated, (x i ,y i ,z i ) is the i-th point closest to the point to be interpolated (x0, y0, z0), is the weight, is the distance between two points, α is the adjustment parameter, i=1,2,…,n, and n is the number of points closest to the interpolation point (x0, y0, z0).
[0093] This embodiment takes the flume test section as an example to reconstruct the riverbed terrain in three dimensions. The specific process is as follows:
[0094] First, a total of 10 black and white checkerboard blocks are placed on both sides of the side wall of the water tank as calibration points, that is, terrain reconstruction control points. A smartphone is used to collect images around the scour pit, with a photo overlap of more than 80%. Overexposure or underexposure of the image during the shooting process will affect the accuracy of point cloud reconstruction. Therefore, a dark or bright shooting environment can ensure a high image recognition matching rate, which can reduce or eliminate the interference caused by moving shadows and static light sources. The total number of photos for each test condition is not less than 200. After each test, there are 280 photos containing black and white checkerboard calibration points, and 10 points are calibrated 280 times.
[0095] Then, the image sequence acquired by the smartphone is 3D reconstructed based on the SFM method, and the sparse point cloud reconstruction image is obtained as shown in the figure. Figure 6 As shown, the dense point cloud reconstruction is shown in Figure 7 As shown, the dense point cloud data is as follows Figure 8 As shown, the point cloud reconstruction diagram is as follows Fig. 9 shown.
[0096] Next, this example sets the study area to select the flow direction range [X 1min , X 1max ]=[1.2,2.4],water flow range [Y 1min , Y 1max]=[0,0.4],unit is m. Fine grid division is performed on the research area, and the physical scale of the grid is set to 1mm×1mm (the physical scale of the grid can be further subdivided according to the experimental conditions).
[0097] Finally, the control points are inserted to obtain the actual terrain three-dimensional coordinates, and the three-dimensional coordinates are interpolated on the bed surface grid, such as Fig.10 As shown, the three-dimensional riverbed topography is obtained. Since the true value of the measurement cannot be obtained, the cross-section elevation can be measured by a laser rangefinder as a detection value after each test, and the longitudinal elevation value can be calculated by the method described in the present invention and compared. The comparison results are shown in Fig.11 As shown in the figure, the dark blue line and the black line represent the elevation values calculated twice by the method of the present invention for the same image sequence, and the green circle points, the red circle points and the sky blue circle points represent the measurement values of the laser rangefinder three times. It can be seen from the figure that the fitting degree of the cross-section elevation is good, which shows that the terrain data carried by the reconstructed point cloud data is very accurate and can accurately obtain the changes of the entire bed surface before and after scouring and silting.
[0098] The technical solution provided by the present invention is described in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, the present invention can also be improved and modified in a number of ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.
Claims
1. A river engineering model test topographic measurement method based on SFM, characterized in that: The steps include: Step 1, obtaining an image sequence of the river engineering model test terrain to be tested; Step 2: Use the SFM method to extract and match the feature points of the acquired image sequence, and reconstruct the sparse point cloud of the image sequence. Then use the CMVS and PMVS methods to reconstruct the dense point cloud of the sparse reconstruction results to achieve three-dimensional reconstruction of the terrain of the river engineering model test to be tested. Step 3: Remove the point cloud data outside the research area from the reconstructed 3D point cloud data, and perform fine grid division on the research area; Step 4: insert control points to obtain the actual three-dimensional coordinates of the terrain of the river engineering model test terrain to be tested, and interpolate the three-dimensional coordinates to the bed grid to obtain a three-dimensional river engineering model that meets the boundary constraints.
2. The SFM-based river engineering model test topographic measurement method according to claim 1 is characterized in that: When acquiring the image sequence in step 1, a plurality of calibration points are respectively set on both sides of the river engineering model test terrain to be measured.
3. The SFM-based river engineering model test topographic measurement method according to claim 2 is characterized in that: The calibration points are black and white checkerboard squares.
4. The SFM-based river engineering model test topographic measurement method according to claim 3 is characterized in that: In the image sequence obtained in step 1, the overlap of images is not less than 80%, and the total number of images is not less than 200.
5. The SFM-based river engineering model test topographic measurement method according to any one of claims 1 to 4, characterized in that: In step 1, a smartphone or digital camera is used to acquire an image sequence.
6. The SFM-based river engineering model test topographic measurement method according to claim 1 is characterized in that: In step 2, the SFM method is used to extract and match the feature points of the acquired image sequence to achieve the three-dimensional reconstruction of the river engineering model test terrain to be tested. The specific steps are as follows: Step 2.1, input the acquired image sequence, and use SIFT operator and SURF operator to detect and extract feature points of each image in the image sequence; Step 2.2, using the KD tree proximity search algorithm, the feature points extracted from different images are matched to obtain feature points with the same name; Step 2.3, according to the two-dimensional coordinates of the camera calibration points and the related geometric constraint relationship equations, calculate the external camera coordinates, direction angles and three-dimensional spatial coordinates; Step 2.4, reconstruct the image sparse point cloud, and expand the coefficient point cloud through bundle adjustment; Step 2.5: Cluster the sparsely reconstructed point cloud data using the CMVS method, and then use the PMVS method to generate a dense point cloud through matching, diffusion, and filtering under the constraints of local photometric consistency and global visibility, complete the dense point cloud reconstruction, and realize the three-dimensional reconstruction of the river engineering model test terrain to be tested.
7. The SFM-based river engineering model test topographic measurement method according to claim 6 is characterized in that: In step 2.2, before feature point matching, the number of feature points of each image is compared according to a preset threshold, images with feature point numbers greater than the preset threshold are matched, and images with feature point numbers less than the preset threshold are deleted.
8. The SFM-based river engineering model test topographic measurement method according to claim 6 is characterized in that: The calculation formula for calculating the external camera coordinates in step 2.3 is: mq=A[RT]P, in, is the camera’s intrinsic parameter matrix, m is the scale correction factor, q=(u,v,1) T are the coordinates in the image coordinate system, (u0, v0) are the coordinates of the projection center in the image plane, (α, β) are the scale parameters in the u and v directions respectively, c is the pixel distortion parameter, P is the terrain coordinate in the world coordinate system, and R and T are the external parameters of the camera.
9. The SFM-based river engineering model test topographic measurement method according to claim 1, characterized in that: The interpolation result expression of the three-dimensional coordinates interpolated onto the bed surface grid in step 4 is as follows: Among them, (x0, y0, z0) represents the point to be interpolated, (x i ,y i ,z i ) is the i-th point closest to the point to be interpolated (x0, y0, z0), is the weight, is the distance between two points, α is the adjustment parameter, i=1,2,…,n, and n is the number of points closest to the interpolation point (x0, y0, z0).
10. The SFM-based river engineering model test topographic measurement method according to claim 1, characterized in that: The boundary constraints include: Continuity constraint: The interpolation result has no discontinuous jumps at the boundary and in the entire area; Smoothness constraint: The curve or surface of the interpolation result is smooth, without sharp corners or drastic fluctuations.
Citation Information
Cited By
Automatic scanning and measuring device and method for scouring and silting terrain of large-scale wave water tank
CN119197469A
A large-scale wave flume automatic scanning device and method for scouring and silting terrain
CN119197469B
