Method for rapidly extracting imaging parameters of reinforcement cage based on Hough clustering

Abstract

The invention relates to the technical field of reinforcement cage detection, in particular to a reinforcement cage imaging parameter rapid extraction method based on Hough clustering. Steel reinforcement cage point cloud data are collected, preprocessed and converted into a two-dimensional form to obtain a steel reinforcement cage image; performing morphological processing on the reinforcement cage image, and classifying the reinforcement cage image after morphological processing by using a method of combining Hough clustering with multi-dimensional feature fusion to obtain a pixel body corresponding to each component of the reinforcement cage; mapping the pixel body corresponding to each part of the reinforcement cage to the three-dimensional point cloud by using a pre-constructed acceleration extraction structure to obtain the three-dimensional point cloud corresponding to each structure of the reinforcement cage; and extracting parameters of each part of the reinforcement cage according to the three-dimensional point cloud corresponding to each structure of the reinforcement cage. According to the rapid extraction method for the imaging parameters of the reinforcement cage, each part in the reinforcement cage can be rapidly identified and extracted, and the parameter extraction efficiency of the reinforcement cage is remarkably improved by accurately classifying and analyzing the structural characteristics of the reinforcement cage.

Description

Technical Field

[0001] The invention relates to the technical field of steel cage detection, and in particular to a method for quickly extracting steel cage image parameters based on Hough clustering. Background Art

[0002] In modern construction projects, the quality and precision of steel cages, as key load-bearing and structural components, directly affect the safety and stability of the building. Especially during the project supervision process, it is crucial to monitor the parameters of the steel cages in real time and accurately.

[0003] Traditional steel cage inspection methods mainly rely on manual measurement. The current steel cage parameter extraction work usually includes the measurement of key parameters such as main reinforcement spacing and spiral reinforcement spacing. However, the manual measurement process is not only time-consuming and labor-intensive, but also easily affected by human factors, resulting in measurement errors, which makes it difficult to meet the project supervision requirements for efficiency and accuracy. This situation makes it impossible for project supervision to accurately grasp the construction quality of the steel cage in real time, which in turn increases the potential safety hazards of buildings and may even cause serious engineering accidents.

[0004] With the continuous development of construction engineering technology, automated detection and parameter extraction methods have gradually become a trend in the field of steel cage detection. Traditional 3D point cloud scanning technology can obtain the precise 3D coordinate data of the steel cage through laser scanners or other sensors, and generate a 3D point cloud containing a large number of points for geometric analysis and fine measurement. However, although this method has the advantage of high precision, its data processing and analysis process is relatively complex and it is difficult to meet the requirements of engineering supervision for rapid response and real-time monitoring. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention provides a method for quickly extracting parameters of steel cage images based on Hough clustering. The method comprises: collecting steel cage point cloud data and converting it into a two-dimensional form after preprocessing to obtain a steel cage image; performing morphological processing on the steel cage image, and classifying the steel cage image after morphological processing by using Hough clustering combined with a multi-dimensional feature fusion method to obtain pixel bodies corresponding to each component of the steel cage; using a pre-constructed acceleration extraction structure to map the pixel bodies corresponding to each component of the steel cage to a three-dimensional point cloud, and obtain a three-dimensional point cloud corresponding to each structure of the steel cage; extracting the parameters of each component of the steel cage according to the three-dimensional point cloud corresponding to each structure of the steel cage. The method for quickly extracting parameters of steel cage images proposed by the present invention can quickly identify and extract each component in the steel cage, and significantly improves the efficiency of steel cage parameter extraction by accurately classifying and analyzing the structural characteristics of the steel cage.

[0006] The present invention adopts the following technical scheme, a method for quickly extracting parameters of a steel cage image based on Hough clustering, comprising:

[0007] Collect and pre-process the steel cage point cloud data;

[0008] The pre-processed steel cage point cloud data is converted into a two-dimensional form to obtain a steel cage image;

[0009] The steel cage image is morphologically processed, and the Hough clustering combined with multi-dimensional feature fusion method is used to classify the steel cage image after morphological processing to obtain the pixel bodies corresponding to each component of the steel cage;

[0010] The pixel bodies corresponding to each component of the steel cage are mapped to a three-dimensional point cloud using a pre-built acceleration extraction structure, thereby obtaining a three-dimensional point cloud corresponding to each structure of the steel cage;

[0011] The parameters of each component of the steel cage are extracted according to the three-dimensional point cloud corresponding to each structure of the steel cage.

[0012] Furthermore, the method for collecting and preprocessing the steel cage point cloud data is as follows:

[0013] Using a statistical outlier filtering method to remove outliers from the steel cage point cloud data;

[0014] A regional threshold is set to select ground points in the steel cage point cloud data, and irrelevant ground points in the steel cage point cloud data are removed using a random sampling consensus algorithm.

[0015] Furthermore, after preprocessing the steel cage point cloud data, it also includes:

[0016] Regularization is performed on the preprocessed steel cage point cloud data. The specific steps are as follows:

[0017] Use the principal component analysis algorithm to calculate the main directions of the point cloud data and align the coordinate axes;

[0018] Calculate the covariance matrix of the point cloud and perform eigenvalue decomposition to obtain eigenvalues ​​and eigenvectors;

[0019] The matrix constructed by the eigenvector is used as the rotation matrix to rotate the point cloud to align it to the main axis, thus completing the regularization of the pre-processed reinforcement cage point cloud data.

[0020] Furthermore, the method for morphological processing of the steel cage image is as follows:

[0021] Assume that the steel cage image is I(x,y), the structural element is B, and the steel cage image after corrosion operation is I eroded (x,y), the corrosion process is expressed as:

[0022]

[0023] The morphological expansion operation is performed on the steel cage image after the corrosion operation to construct the structural element C. The expansion operation result image is I dilated (x,y), the expansion operation is expressed as:

[0024]

[0025] Among them, u represents the pixel index value of the image's horizontal axis, and v represents the pixel index value of the image's vertical axis.

[0026] Furthermore, the Hough clustering combined with multi-dimensional feature fusion method is used to classify the steel cage images after morphological processing, including:

[0027] The Canny operator is used to detect the edge of the steel cage image after morphological processing;

[0028] The steel cage image after edge detection is converted into Hough space to obtain the pixel distribution of the steel cage image in Hough space;

[0029] The pixel distribution is combined into a multi-dimensional feature vector, and a clustering algorithm based on the feature vector is used to cluster the multi-dimensional feature vector to obtain various components of the steel cage image.

[0030] Further, the pixel distribution is combined into a multi-dimensional feature vector, and a feature vector-based clustering algorithm is used to cluster the multi-dimensional feature vector, including:

[0031] According to the pixel distribution of the steel cage image in the Hough space, construct the output line segments in the steel cage image;

[0032] The multidimensional feature vector is constructed according to the output line segment: Among them, ρ i is the polar coordinate of the i-th output line segment in Hough space, θ i is the direction angle of the i-th output line segment, l i is the length of the i-th output line segment, are the horizontal and vertical coordinates of the center point of the i-th output line segment;

[0033] Weighted K-means clustering is used to perform weighted clustering on each component in the multidimensional feature vector.

[0034] Furthermore, the pixel volumes corresponding to the various components of the steel cage are mapped to the three-dimensional point cloud using the pre-built acceleration extraction structure, specifically:

[0035] The KD tree construction method is used to map the pixel volumes corresponding to each component of the steel cage to the three-dimensional point cloud.

[0036] Furthermore, after extracting the parameters of each component of the steel cage, it also includes:

[0037] The parameters of each component of the steel cage are compared with the pre-set standard parameters in Json format within the allowable error.

[0038] The beneficial effect of the present invention is that the present invention converts the point cloud data of the steel cage into an image form and combines the line segment detection and clustering technology in the Hough transform. The method can quickly identify and extract the main bars, spiral bars and bundled bars in the steel cage. In particular, by utilizing the scattered point distribution characteristics in the Hough transform process and combining the geometric characteristics of the steel bars such as the specific angle, the structural characteristics of the steel cage can be accurately classified and analyzed. Not only the speed advantage of image processing is fully utilized, but also the accuracy advantage of point cloud data is combined, thereby significantly improving the efficiency of steel cage parameter extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 A schematic diagram of a method for quickly extracting parameters of a steel cage image based on Hough clustering according to an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of comparison before and after removal of ground points of a steel cage data according to an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of pixel distribution of a steel cage image after corrosion and expansion in Hough space according to an embodiment of the present invention;

[0043] Figure 4 A schematic diagram of a result of multi-dimensional feature fusion clustering analysis of a steel cage image according to an embodiment of the present invention;

[0044] Figure 5 This is a schematic diagram of main reinforcement position obtained by morphological analysis of a steel cage side view in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] A schematic diagram of a method for quickly extracting parameters of a steel cage image based on Hough clustering according to an embodiment of the present invention is shown in FIG. Figure 1 As shown, including:

[0047] Data preprocessing: First, by removing outliers and noise in the point cloud, the quality of the data can be improved, providing a basis for subsequent processing. Secondly, ground points in the point cloud data are removed by ground point removal to reduce the interference of related background data and focus on processing the effective point cloud of the steel cage; at the same time, the point cloud data is regularized to ensure that the steel cage point cloud data is calculated in a unified coordinate system to improve the accuracy and consistency of the calculation.

[0048] Two-dimensional image morphological operation preprocessing: In order to realize the rapid extraction of relevant parameters of the steel cage during the supervision process, the present invention chooses to project the regularized point cloud data onto a two-dimensional plane and convert it into an image form. Subsequently, corrosion and expansion operations are used to perform morphological processing on the image to enhance the contour of the steel bars and remove small noise and gaps in the image. The raster voxelization processing accelerates the point cloud data processing while improving the robustness of the steel bar component classification.

[0049] Hough clustering division: for the image after morphological preprocessing, the present invention combines the multidimensional feature fusion method in the Hough space, fuses the corresponding line segment scatter point features with geometric features such as angle, position distribution and length, and realizes automatic classification of the detected line segments through the clustering algorithm, and extracts the corresponding geometric parameters at the same time.

[0050] 2D to 3D recovery and extraction: In order to correspond the 2D voxel results obtained in Hough clustering to 3D space, the present invention pre-constructs a corresponding accelerated extraction structure; based on the mapping relationship between point cloud data and image, the structure can efficiently restore the clustering results from the 2D image coordinate system to the original 3D point cloud coordinate system, thereby realizing accurate extraction of various types of steel bar point clouds;

[0051] Structural analysis and calculation: After completing the three-dimensional extraction, the classified steel bar types are subjected to structural analysis, and relevant parameters of the steel cage are calculated, such as the diameter of the steel cage, the spacing between the spiral bars and the main bars, etc. Based on these parameters, the present invention can automatically generate a test report, providing fast and reliable data support for project supervision and construction quality control.

[0052] In another specific embodiment of the present invention, the method for quickly extracting steel cage image parameters of the present invention specifically includes:

[0053] Collect and pre-process the steel cage point cloud data;

[0054] In the preprocessing stage, the embodiment of the present invention adopts a statistical outlier filtering method to preprocess the point cloud of the steel cage to remove a small number of other miscellaneous points outside the main body of the steel cage, and determines whether it is an outlier by calculating the distance between each point and its neighboring points. The specific steps are as follows: for each point in the point cloud, calculate its Euclidean distance to the k nearest neighboring points; calculate the mean and standard deviation of these distances; set a distance threshold, which is usually the mean plus several times the standard deviation, and points exceeding the threshold are considered to be outliers.

[0055] Set point p i The neighborhood point set is N i , contains k points, the mean distance is μ i , with standard deviation σ i , the distance threshold T is defined as:

[0056] T=μ i +ασ i

[0057] Here, α is a user-defined parameter, usually ranging from 1 to 3.

[0058] Point p i The average distance μ to its neighboring points i and standard deviation σ i The calculation is as follows:

[0059]

[0060] Among them, d(p i ,p j ) represents point p i and the neighborhood point p j The Euclidean distance between .

[0061] Further, the embodiment of the present invention adopts a random sampling consensus (RANSAC) method, sets a regional threshold, selects a range of ground points from all the collected points, fits the plane equation multiple times and peels off the ground points one by one, and removes irrelevant ground points from the steel cage point cloud data. The comparison diagram before and after removal is shown in FIG. Figure 2 As shown, it provides a basis for subsequent image morphological operations, specifically:

[0062] The coordinate system of the original collected point cloud data is not regularized, and the point cloud needs to be regularized to facilitate the projection and other steps under the subsequent image morphological operation; the embodiment of the present invention moves the centroid of the point cloud to the origin, and assumes that the point in the point cloud data is P i =(x i ,y i ,z i ), the centroid is Translate each point in the point cloud data to the new coordinate system: After alignment to the main axis, principal component analysis (PCA) is used to calculate the main directions of the point cloud data and align the coordinate axes. The covariance matrix C of the point cloud is calculated: Perform eigenvalue decomposition to obtain eigenvalues ​​and eigenvectors: Where Λ is a diagonal matrix, the diagonal is the eigenvalue, V is the eigenvector, and then the eigenvector matrix is ​​used as the rotation matrix to rotate the point cloud to the main axis: This completes the regularization of the point cloud data.

[0063] The pre-processed steel cage point cloud data is converted into a two-dimensional form to obtain a steel cage image;

[0064] In order to realize the rapid extraction of relevant parameters of the steel cage during the supervision process, the embodiment of the present invention performs XY plane projection on the regularized point cloud, sets the voxel grid size to generate the corresponding binary image, and there may still be some small objects and noise in the image. For robust processing, the morphological corrosion operation is first used to make the remaining objects more coherent. Specifically, the corrosion operation will reduce the boundary of the foreground (white) area in the image and expand the background (black) area, so as to reduce the influence of noise points on subsequent detection.

[0065] The steel cage image is morphologically processed, and the Hough clustering combined with multi-dimensional feature fusion method is used to classify the steel cage image after morphological processing to obtain the pixel bodies corresponding to each component of the steel cage;

[0066] Suppose the input image after projection is I(x,y), the structural element is B, and the erosion operation image is I eroded (x,y), the corrosion process is defined as:

[0067]

[0068] The morphological expansion operation is performed on the steel cage image after the corrosion operation to construct the structural element C. The expansion operation result image is I dilated (x,y), the expansion operation is expressed as:

[0069]

[0070] Among them, u represents the pixel index value of the image's horizontal axis, and v represents the pixel index value of the image's vertical axis. The dilation operation can enhance the contour of the target object, fill small holes, connect broken areas, and make the target object more coherent.

[0071] Perform line segment detection on the image after erosion and expansion operations. Before using Hough transform line segment detection, first perform Canny edge detection, calculate the image gradient of the input image and obtain the detection result E(x,y):

[0072]

[0073] Gradient magnitude and direction:

[0074]

[0075] Then the image pixels are converted to the parameter space, namely the Hough space, for point detection in the parameter space, using the linear polar coordinate representation in the Hough space:

[0076] ρ=xcosθ+ysinθ

[0077] Among them, ρ is the vertical distance from the origin to the straight line, θ is the angle between the vertical distance and the x-axis. For each image point (edge ​​point), calculate all possible (ρ, θ) pairs corresponding to it, and obtain the scatter plot corresponding to (ρ, θ) under the corresponding accumulator as follows Figure 3 As shown, further weighted cluster analysis is performed to reduce the reliance on artificial parameter settings in the process of steel cage component extraction and improve the robustness of the method.

[0078] Multidimensional feature fusion and cluster analysis, after completing the preliminary (ρ, θ) distribution voting in Hough space, the present invention further introduces multidimensional feature fusion and cluster analysis to realize the automatic classification of steel bar types. Taking into account the geometric characteristics of spiral bars, main bars, and bundled bars, the present invention considers the line segment length l for each line segment corresponding to (ρ, θ) i , line segment center position Segment direction angle θ i To construct the multi-dimensional features of each final output line segment, i.e. each steel bar component. The values ​​of each component are calculated as follows:

[0079]

[0080] The embodiment of the present invention combines the above features into a multi-dimensional feature vector for subsequent cluster analysis; for each line segment L i , construct its feature vector v i as follows:

[0081]

[0082] Then, a clustering algorithm based on feature vectors is used to classify the line segments. In the embodiment of the present invention, weighted K-means clustering is used to assign different weights to each component in the feature vector to meet the specific needs of steel bar detection. Let w = [w ρ ,w θ ,w l ,w x ,w y] is the weight vector, where the main weight reference is the angle weight, which can quickly realize classification, and the weighted feature vector is:

[0083]

[0084] The line segments are clustered by the weighted K-means algorithm to achieve the classification of main reinforcement, bundled reinforcement and spiral reinforcement. The results are further optimized to ensure the accuracy of classification by removing noise segments and segments that do not meet geometric characteristics. Figure 4 As shown, the inclined angle is spiral reinforcement, the vertical is bundled reinforcement, and the horizontal is bundled reinforcement. The final output is the detection results of various steel bar components in the steel cage, including their key parameters such as position, angle and length in the two-dimensional image, and prepare to provide a basis for the next step of two-dimensional to three-dimensional restoration and extraction.

[0085] It should be noted that, when the image of the steel cage is a side view, the embodiment of the present invention also uses the expansion operation followed by the morphological connected domain analysis to locate the positions of the main bars. Figure 5 As shown, it can be used for automated standard comparison in subsequent steps.

[0086] The pixel bodies corresponding to each component of the steel cage are mapped to a three-dimensional point cloud using a pre-built acceleration extraction structure, thereby obtaining a three-dimensional point cloud corresponding to each structure of the steel cage;

[0087] The embodiment of the present invention maps the two-dimensional line segment results obtained by Hough transform and cluster analysis in the previous step back to three-dimensional space through two-dimensional to three-dimensional recovery extraction, thereby realizing accurate extraction of different types of steel bars in the steel cage. By making one-to-one correspondence between the detected line segment index and the three-dimensional point cloud index, the entire steel bar extraction process does not require manual intervention and is fully automated, thereby improving detection efficiency and accuracy.

[0088] According to the three-dimensional point cloud index corresponding to the pixel voxels of the spiral reinforcement, main reinforcement, and binding reinforcement components output after multi-dimensional feature clustering, the embodiment of the present invention uses a pre-built acceleration extraction structure to complete the mapping from the two-dimensional pixel voxel results to the three-dimensional point cloud. By pre-processing the input point cloud, an acceleration structure based on space division is constructed. One method of the embodiment of the present invention adopts a KD tree structure, specifically: for each point P (x, y, z) in the point cloud, it is inserted into the KD tree in sequence according to its spatial position, and all points in the point cloud are divided into multiple areas. In the process of three-dimensional point cloud recovery, the point cloud area corresponding to the line segment is quickly found to reduce the mapping time.

[0089] The parameters of each component of the steel cage are extracted according to the three-dimensional point cloud corresponding to each structure of the steel cage.

[0090] The embodiment of the present invention realizes 3D point cloud extraction by using an accelerated extraction structure to map the 2D pixel voxel result to the index of the 3D point cloud. i (x, y) uses the 3D KD tree index in the acceleration structure to quickly find the corresponding 3D point cloud P(x, y, z). At the same time, in order to ensure the integrity of the extracted 3D point cloud, the neighborhood radius r is increased for the 3D point cloud of the line segment during the mapping process, and the search range δr is expanded to ensure that all relevant point clouds are completely extracted. Finally, the point cloud size of each part and other supervision-related data are output, such as the diameter D of the steel cage, the main reinforcement spacing S, etc., so as to realize the rapid extraction of the parameters of the steel cage components.

[0091] Finally, for the measurement results of the divided main reinforcement, spiral reinforcement, and bundled reinforcement, the embodiment of the present invention further compares them with the pre-set Json format standard parameters to determine the allowable errors of each design structure, thereby realizing full-process automated measurement and outputting the final steel cage design qualification report to quickly complete the corresponding supervision work.

[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for quickly extracting parameters of steel cage images based on Hough clustering, characterized in that: include: Collect and pre-process the steel cage point cloud data; The pre-processed steel cage point cloud data is converted into a two-dimensional form to obtain a steel cage image; The steel cage image is morphologically processed, and the Hough clustering combined with multi-dimensional feature fusion method is used to classify the steel cage image after morphological processing to obtain the pixel bodies corresponding to each component of the steel cage; The pixel bodies corresponding to each component of the steel cage are mapped to a three-dimensional point cloud using a pre-built acceleration extraction structure, thereby obtaining a three-dimensional point cloud corresponding to each structure of the steel cage; The parameters of each component of the steel cage are extracted according to the three-dimensional point cloud corresponding to each structure of the steel cage.

2. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: The method for collecting and preprocessing the steel cage point cloud data is as follows: Using a statistical outlier filtering method to remove outliers from the steel cage point cloud data; A regional threshold is set to select ground points in the steel cage point cloud data, and irrelevant ground points in the steel cage point cloud data are removed using a random sampling consensus algorithm.

3. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: After preprocessing the steel cage point cloud data, it also includes: Regularization is performed on the preprocessed steel cage point cloud data. The specific steps are as follows: Use the principal component analysis algorithm to calculate the main directions of the point cloud data and align the coordinate axes; Calculate the covariance matrix of the point cloud and perform eigenvalue decomposition to obtain eigenvalues ​​and eigenvectors; The matrix constructed by the eigenvector is used as the rotation matrix to rotate the point cloud to align it to the main axis, thus completing the regularization of the pre-processed reinforcement cage point cloud data.

4. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: The method for morphological processing of steel cage images is: Assume that the steel cage image is I(x,y), the structural element is B, and the steel cage image after corrosion operation is I eroded (x,y), the corrosion process is expressed as: The morphological expansion operation is performed on the steel cage image after the corrosion operation to construct the structural element C. The expansion operation result image is I dilated (x,y), the expansion operation is expressed as: Among them, u represents the pixel index value of the image's horizontal axis, and v represents the pixel index value of the image's vertical axis.

5. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: The Hough clustering combined with multi-dimensional feature fusion method is used to classify the steel cage image after morphological processing, including: The Canny operator is used to detect the edge of the steel cage image after morphological processing; The steel cage image after edge detection is converted into Hough space to obtain the pixel distribution of the steel cage image in Hough space; The pixel distribution is combined into a multi-dimensional feature vector, and a clustering algorithm based on the feature vector is used to cluster the multi-dimensional feature vector to obtain various components of the steel cage image.

6. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 5 is characterized in that: Combining the pixel distribution into a multi-dimensional feature vector, and clustering the multi-dimensional feature vector using a feature vector-based clustering algorithm, including: According to the pixel distribution of the steel cage image in the Hough space, construct the output line segments in the steel cage image; The multidimensional feature vector is constructed according to the output line segment: Among them, ρ i is the polar coordinate of the i-th output line segment in Hough space, θ i is the direction angle of the i-th output line segment, l i is the length of the i-th output line segment, are the horizontal and vertical coordinates of the center point of the i-th output line segment; Weighted K-means clustering is used to perform weighted clustering on each component in the multidimensional feature vector.

7. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: The pre-built acceleration extraction structure is used to map the pixel volumes corresponding to each component of the steel cage to a 3D point cloud, specifically: The KD tree construction method is used to map the pixel volumes corresponding to each component of the steel cage to the three-dimensional point cloud.

8. The method for rapid extraction of steel cage image parameters based on Hough clustering according to claim 1 is characterized in that: After extracting the parameters of each component of the steel cage, it also includes: The parameters of each component of the steel cage are compared with the pre-set standard parameters in Json format within the allowable error.

Images