A method, device and medium for detecting the quality of a concrete segment

Abstract

The embodiment of the specification discloses a segment quality detection method and device of a concrete segment, and a medium, relates to the technical field of point cloud data processing, and the method comprises the following steps: acquiring segment scanning point cloud data corresponding to a to-be-detected concrete segment and segment standard part point cloud data corresponding to the to-be-detected concrete segment; respectively extracting curved surface point clouds of the segment scanning point cloud data and the segment standard part point cloud data, and determining a scanning curved surface point cloud set and a standard curved surface point cloud set; determining a curved surface corresponding relationship of each scanning curved surface point cloud and the standard curved surface point cloud according to the scanning curved surface point cloud set and the standard curved surface point cloud set, so as to determine a plurality of curved surface transformation matrices based on the curved surface corresponding relationship; registering the segment scanning point cloud data and the segment standard part point cloud data through the plurality of curved surface transformation matrices, determining segment registration point cloud data, and detecting the segment quality of the to-be-detected concrete segment based on the segment registration point cloud data.

Description

Technical Field

[0001] This specification relates to the field of point cloud data processing technology, and in particular to a method, equipment and medium for inspecting the quality of concrete pipe segments. Background Technology

[0002] Pre-assembly of steel structures is commonly seen in applications such as large steel bridges, wind turbine blades, and power towers. For steel bridge pre-assembly technology, the process involves horizontal assembly in the factory, horizontal assembly on-site after painting, and finally, on-site hoisting and erection. Horizontal assembly and hoisting are labor-intensive and require the use of total stations and levels. If the bridge span is large, a large site support is needed to support the horizontal assembly process. Therefore, the industry's desire for digital simulation pre-assembly is growing stronger. In the pre-assembly stage, the factory quality and precision of concrete segments play a crucial role, affecting not only the safety and stability of structures such as wind turbine towers but also the long-term operating efficiency and safety of wind turbines. Currently, when inspecting the quality of concrete segments, point cloud registration technology is used to align them with standard segments for dimensional comparison, ensuring that both are compared in the same coordinate system. Traditional point cloud registration methods focus on calculating statistically significant features and then solving for the required transformation matrix (including rotation and translation parameters) by matching these features. If point clouds are missing, it will severely interfere with various statistically significant features, resulting in inaccurate registration results.

[0003] For the large concrete segments unique to wind turbine towers, each segment is enormous, exceeding four meters in height and with a maximum arc length exceeding ten meters. Furthermore, their weight ranges from three to six tons, making it impossible to accurately obtain segment dimensions manually. Additionally, after leaving the factory, the segments are only permitted to be placed, stored, and transported upright, resulting in incomplete and missing data from measurements and scans taken in a vertical position. Therefore, the acquisition of point clouds for the concrete segments is incomplete, leading to inaccurate registration results from traditional point cloud matching methods, and consequently affecting the quality inspection results of the segments. Summary of the Invention

[0004] This specification provides one or more embodiments of a method, equipment, and medium for inspecting the quality of concrete pipe segments, which addresses the following technical problem: the acquisition of point clouds of concrete pipe segments is incomplete, leading to inaccurate registration results from traditional point cloud matching methods, thereby affecting the quality inspection results of the pipe segments.

[0005] One or more embodiments of this specification employ the following technical solutions:

[0006] This specification provides one or more embodiments of a method for inspecting the quality of concrete pipe segments. The method includes: acquiring scanned point cloud data of the concrete pipe segment to be inspected and point cloud data of a standard component of the pipe segment to be inspected; extracting surface point clouds from the scanned point cloud data and the standard component point cloud data to determine a scanned surface point cloud set and a standard surface point cloud set, wherein the scanned surface point cloud set includes multiple scanned surface point clouds and the standard surface point cloud set includes multiple standard surface point clouds; determining the surface correspondence relationship of each scanned surface point cloud and the standard surface point cloud based on the scanned surface point cloud set and the standard surface point cloud set, and determining multiple corresponding surface transformation matrices based on the surface correspondence relationship; registering the scanned point cloud data of the pipe segment with the point cloud data of the standard component of the pipe segment using the multiple surface transformation matrices to determine the registered point cloud data of the pipe segment, and performing pipe segment quality inspection on the concrete pipe segment to be inspected based on the registered point cloud data of the pipe segment.

[0007] Further, surface point cloud extraction is performed on the scanned point cloud data of the tunnel segment and the point cloud data of the standard tunnel segment, respectively, to determine the scanned surface point cloud set and the standard surface point cloud set. Specifically, this includes: analyzing each scanned point cloud in the scanned point cloud data of the tunnel segment and each standard point cloud in the point cloud data of the standard tunnel segment to determine the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud. The scanned point cloud feature information includes scanned point distance information and scanned normal vector information, and the standard point cloud feature information includes standard point distance information and standard normal vector information. Based on the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud, the scanned surface point cloud set and the standard surface point cloud set are determined.

[0008] Further, based on the scan point cloud feature information of each scan point cloud and the standard point cloud feature information of each standard point cloud, a scanned surface point cloud set and a standard surface point cloud set are determined. Specifically, this includes: based on the scan point distance information of each scan point cloud and the standard point distance information of each standard point cloud, noise points are removed from the tube segment scan point cloud data and the tube segment standard part point cloud data, respectively, to determine the first tube segment scan point cloud data and the first tube segment standard part point cloud data; based on the scan normal vector information of each scan point cloud, multiple target scanned surface point clouds in the first tube segment scan point cloud data are determined, and the number of scan points in each target scanned surface point cloud is obtained; and the standard point distance information of each standard point cloud is used to determine the target scanned surface point cloud data. Using normal vector information, multiple target standard surface point clouds are determined in the point cloud data of the first segment standard component, and the number of standard points in each target standard surface point cloud is obtained. Based on a preset first point count threshold and the number of scan points in each target scan surface point cloud, the multiple target scan surface point clouds are filtered to determine the multiple scan surface point clouds, wherein the number of scan points in each scan surface point cloud is not less than the first point count threshold. Based on a second point count threshold and the number of standard points in each target standard surface point cloud, the multiple target standard surface point clouds are filtered to determine the multiple standard surface point clouds, wherein the number of standard points in each standard surface point cloud is not less than the second point count threshold.

[0009] Further, based on the scanned surface point cloud set and the standard surface point cloud set, the surface correspondence relationship of each scanned surface point cloud and the standard surface point cloud is determined, specifically including: obtaining multiple scanned surface point clouds in the scanned surface point cloud set and multiple standard surface point clouds in the standard surface point cloud set; extracting point cloud features for each scanned surface point cloud and each standard surface point cloud to determine the scanned surface features corresponding to each scanned surface point cloud and the standard surface features corresponding to each standard surface point cloud; performing feature matching on each scanned surface point cloud and the standard surface point cloud based on each scanned surface feature and the standard surface feature to determine the specified standard surface point cloud corresponding to each scanned surface point cloud, thereby determining the surface correspondence relationship.

[0010] Further, based on the surface correspondence, determining multiple corresponding surface transformation matrices specifically includes: determining a specified scanned surface point cloud and a corresponding specified standard surface point cloud through the surface correspondence; performing standardization processing on the specified scanned surface point cloud and the specified standard surface point cloud respectively to determine the processed scanned surface point cloud corresponding to the specified scanned surface point cloud and the processed standard surface point cloud corresponding to the specified standard surface point cloud, and determining the standardization transformation matrix and the corresponding standardization transformation inverse matrix; determining the flipped scanned surface point cloud and the flipped transformation matrix based on the processed scanned surface point cloud and the processed standard surface point cloud; performing registration based on the flipped scanned surface point cloud and the processed standard surface point cloud using an iterative nearest-point algorithm to determine the registration transformation matrix; and determining multiple corresponding surface transformation matrices based on the standardization transformation matrix, the standardization transformation inverse matrix, the flipped transformation matrix, and the registration transformation matrix.

[0011] Further, based on the processed scanned surface point cloud and the processed standard surface point cloud, the flipped scanned surface point cloud and the flip transformation matrix are determined. Specifically, this includes: performing a flipping operation on the processed scanned surface point cloud according to a preset flipping method to determine multiple candidate flipped scanned surface point clouds, wherein the flipping method includes rotating around the X-axis by a specified angle, rotating around the Y-axis by a specified angle, and rotating around the Z-axis by a specified angle; matching the multiple candidate flipped scanned surface point clouds and the processed scanned surface point cloud with the processed standard surface point cloud respectively to determine the flipped scanned surface point cloud; and determining the flip transformation matrix based on the flipped scanned surface point cloud.

[0012] Further, the segment scan point cloud data and the segment standard part point cloud data are registered using the multiple surface transformation matrices to determine the segment registered point cloud data. Specifically, this includes: determining the normalization transformation matrix, normalization inverse matrix, flip transformation matrix, and registration transformation matrix in the surface transformation matrix; performing point cloud normalization processing on the segment scan point cloud data using the normalization transformation matrix to determine the normalized segment scan point cloud data; flipping the normalized segment scan point cloud data according to the flip transformation matrix to determine the flipped segment scan point cloud data; performing registration transformation on the flipped segment scan point cloud data based on the registration transformation matrix to determine the registered segment scan point cloud data; and performing transformation processing on the registered segment scan point cloud data using the normalization inverse matrix to determine the segment registered point cloud data.

[0013] Further, based on the segment registration point cloud data, the quality inspection of the concrete segment to be inspected is performed, specifically including: extracting multi-dimensional features from the segment registration point cloud data to determine the corresponding dimensional feature data and geometric feature data; using the dimensional feature data, performing segment size inspection on the concrete segment to be inspected to determine the dimensional deviation data corresponding to the concrete segment to be inspected; matching the geometric feature data with the standard component geometric feature data corresponding to the pre-acquired segment standard component point cloud data to determine the shape consistency matching data corresponding to the concrete segment to be inspected; and determining the quality assessment information corresponding to the concrete segment to be inspected based on the dimensional deviation data and the shape consistency matching data, so as to perform segment quality inspection on the concrete segment to be inspected.

[0014] This specification provides one or more embodiments of a concrete tunnel segment quality inspection device, comprising:

[0015] At least one processor; and,

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the above-described method.

[0018] This specification provides one or more embodiments of a non-volatile computer storage medium storing computer-executable instructions configured to perform the above-described method.

[0019] The above-mentioned at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects: By performing surface point cloud extraction on the scanned point cloud data of the tube segment and the point cloud data of the standard tube segment separately, the actual surface shape of the tube segment can be more accurately focused on, rather than noise or irrelevant parts in the entire point cloud data, reducing registration errors caused by missing data, because surface point cloud extraction can usually retain key shape information while removing unnecessary background or noise; the surface correspondence between each scanned surface point cloud and the standard surface point cloud is determined, and multiple surface variations are calculated based on this correspondence. Matrix replacement is a more refined and accurate registration method. Compared with traditional global registration methods, it can take into account the local differences between different curved surfaces of the tunnel segment, thereby achieving more accurate local registration and reducing registration errors caused by incomplete data or scanning angle limitations. Quality inspection based on the registered point cloud data of the tunnel segment can ensure the accuracy and reliability of the inspection results. Since the registration process has minimized the errors caused by missing data and inaccurate registration, the extracted size feature data and geometric feature data will be closer to the real situation, improving the accuracy and reliability of quality inspection. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0021] Figure 1 A schematic flowchart illustrating a method for inspecting the quality of concrete tunnel segments provided in this specification.

[0022] Figure 2 This is a structural schematic diagram of a concrete pipe segment quality inspection device provided in the embodiments of this specification. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0024] This specification provides a method for inspecting the quality of concrete pipe segments. It should be noted that the executing entity in this specification can be a server or any device with data processing capabilities. Figure 1 This is a flowchart illustrating a method for inspecting the quality of concrete tunnel segments provided in an embodiment of this specification. Figure 1 As shown, the main steps include the following:

[0025] Step S101: Obtain the point cloud data of the concrete segment to be inspected and the point cloud data of the standard component of the concrete segment to be inspected.

[0026] In one embodiment of this specification, the scanning equipment is determined based on the size, shape, surface material, and detection accuracy requirements of the segment. Generally, a high-precision 3D laser scanner or other suitable scanning equipment is used to collect the point cloud data of the concrete segment to be inspected. During the scanning process, the scanning environment should be clean and tidy, avoiding debris and unnecessary interference. Furthermore, the lighting should be uniform to reduce shadows and reflections. The concrete segment to be inspected is placed in a suitable scanning position, and the scanning equipment is used to perform a full-range scan. During the scanning process, care should be taken to maintain the stability and accuracy of the equipment, avoiding scanning errors caused by shaking and vibration. After scanning, the point cloud data of the segment is exported from the scanning equipment, usually in the form of a point cloud file (such as .las, .pcd, .ply, etc.). To obtain standard part information, it is first necessary to determine the standard size, shape, and material of the segment to be inspected. If existing standard part point cloud data exists, it can be used directly. If standard part point cloud data does not exist, it can be generated using the following method: a 3D model of the standard part is drawn using CAD software, and the 3D model is imported into point cloud generation software to generate the corresponding point cloud data. Alternatively, a high-precision 3D printer can be used to print out standard parts, which can then be scanned to obtain point cloud data.

[0027] Step S102: Extract surface point clouds from the scanned point cloud data of the tube segment and the point cloud data of the standard tube segment to determine the scanned surface point cloud set and the standard surface point cloud set.

[0028] The scanned surface point cloud set includes multiple scanned surface point clouds, and the standard surface point cloud set includes multiple standard surface point clouds.

[0029] Traditional point cloud matching algorithms calculate various statistically significant features and determine the specific transformation matrix by matching these features. However, in cases where the scanned point cloud is missing (e.g., the presence of a ground contact surface during scanning prevents this surface from being scanned), these statistically significant features can be interfered with, leading to inaccurate registration results. In one embodiment of this specification, the point cloud is split into several sets of surfaces. Since matching between surfaces is more accurate and relatively easier than matching the entire point cloud, the overall calculated features can be converted into a surface-to-surface approach. Surface point cloud extraction is performed on the scanned point cloud data of the tunnel segment and the point cloud data of the standard tunnel segment, respectively, to determine a scanned surface point cloud set containing several scanned surface point clouds and a standard surface point cloud set containing several standard surface point clouds.

[0030] Surface point cloud extraction is performed on the scanned point cloud data of the tunnel segment and the point cloud data of the standard part of the tunnel segment to determine the scanned surface point cloud set and the standard surface point cloud set. Specifically, this includes: analyzing each scanned point cloud in the scanned point cloud data of the tunnel segment and each standard point cloud in the point cloud data of the standard part of the tunnel segment to determine the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud. The scanned point cloud feature information includes scanned point distance information and scanned normal vector information, and the standard point cloud feature information includes standard point distance information and standard normal vector information. Based on the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud, the scanned surface point cloud set and the standard surface point cloud set are determined.

[0031] In one embodiment of this specification, for each scanned point cloud in the segment scanning point cloud data, its distance to neighboring points is calculated. This can be achieved by using the K-Nearest Neighbor (K-NN) algorithm to find the K nearest points and calculating the average distance or minimum / maximum distance as the point distance information. A local surface fitting method (such as PCA) is used to estimate the normal vector of each scanned point, finding the local neighborhood around that point, and using these points to fit a local plane or surface. The normal vector of this plane or surface is the normal vector of that point. The extraction methods for standard point distance information and standard normal vector information are similar to those for scanned point clouds, but are applied to the segment standard component point cloud data.

[0032] Using the point distance and normal vector information of the scanned point cloud as features, cluster analysis is performed on the scanned point cloud. Clustering algorithms (such as K-means, DBSCAN, etc.) can divide the point cloud into different surface regions based on these features. During the clustering process, clustering parameters (such as the number of clusters, neighborhood size, etc.) can be adjusted as needed to ensure the accuracy and rationality of the surface division. The point cloud set corresponding to each surface region is extracted from the clustering results, i.e., the scanned surface point cloud set. The method for determining the standard surface point cloud set is similar to that for the scanned surface point cloud set, but it is directly applied to the point cloud data of the standard pipe segment. Since the point cloud data of the standard pipe segment has a clear geometric shape and surface definition, the determination of its surface point cloud set may be more direct and accurate.

[0033] Based on the scan point cloud feature information of each scan point cloud and the standard point cloud feature information of each standard point cloud, the scan surface point cloud set and the standard surface point cloud set are determined. Specifically, this includes: based on the scan point distance information of each scan point cloud and the standard point distance information of each standard point cloud, noise points are removed from the tube segment scan point cloud data and the tube segment standard part point cloud data, respectively, to determine the first tube segment scan point cloud data and the first tube segment standard part point cloud data; based on the scan normal vector information of each scan point cloud, multiple target scan surface point clouds are determined in the first tube segment scan point cloud data, and the number of scan points in each target scan surface point cloud is obtained; and the standard normal vector of each standard point cloud is used to determine the target scan surface point cloud data. Vector information is used to determine multiple target standard surface point clouds in the point cloud data of the first tube segment standard component, and the number of standard points in each target standard surface point cloud is obtained; based on a preset first point count threshold and the number of scan points in each target scan surface point cloud, the multiple target scan surface point clouds are filtered to determine the multiple scan surface point clouds, wherein the number of scan points in each scan surface point cloud is not less than the first point count threshold; according to a second point count threshold and the number of standard points in each target standard surface point cloud, the multiple target standard surface point clouds are filtered to determine the multiple standard surface point clouds, wherein the number of standard points in each standard surface point cloud is not less than the second point count threshold.

[0034] In one embodiment of this specification, each point in the tube segment scan point cloud data is traversed, and its scan point distance information (such as the average distance or standard deviation to neighboring points) is compared with a preset noise threshold. If the point distance information of a certain point exceeds the noise threshold, the point is considered a noise point and is removed from the point cloud data. The above process is repeated until all noise points are removed, resulting in the first tube segment scan point cloud data. The same method is used to remove noise points from the tube segment standard part point cloud data to obtain the first tube segment standard part point cloud data. It should be noted that the noise threshold can be set based on experience from processing tube segment point cloud data in the past, and should be reasonably set according to the characteristics of the actual point cloud data and the detection requirements to avoid erroneously removing valid points or retaining too many noise points.

[0035] Traverse each point in the first segment's scanned point cloud data, dividing the point cloud into different potential surface regions based on its scanned normal vector information (using clustering algorithms or methods based on normal vector similarity). For each potential surface region, calculate the number of scanned points within it. Determine multiple target scanned surface point clouds, representing the main surfaces in the scanned data. Using a similar method, divide the first segment's standard part point cloud data into different potential surface regions based on standard normal vector information. Calculate the number of standard points in each potential surface region. Based on the scanning accuracy of the scanning equipment, preset a first point count threshold, which represents the minimum number of points required for a valid surface point cloud. Traverse all target scanned surface point clouds, checking if the number of scanned points in each point cloud is not less than the point count threshold. Select the scanned surface point clouds that meet the conditions as the final multiple scanned surface point clouds. Based on the scanning parameters or construction parameters during the standard part construction process, set a second point count threshold, and use the second point count threshold to filter the target standard surface point clouds. Select the standard surface point clouds that meet the conditions as the final multiple standard surface point clouds.

[0036] The above technical solutions significantly improve the quality of point cloud data by eliminating noise points. Noise points are usually generated due to random errors during the scanning process, environmental factors, or irregularities on the object's surface. They may interfere with subsequent data analysis and processing. By setting a reasonable noise threshold and eliminating these points, errors can be reduced, making the data more accurate and reliable. Through surface partitioning based on normal vector information and filtering based on point count thresholds, point cloud data representing the main surfaces of the tunnel segment can be extracted. This point cloud data has sufficient points and clear surface features, enabling it to more accurately represent the shape and structure of the tunnel segment. The elimination of noise points and the filtering of invalid surface point clouds can reduce the amount of data that needs to be processed, thereby improving the efficiency of data processing. In subsequent steps, such as surface reconstruction, matching, or comparative analysis, using a more concise and accurate dataset can significantly reduce computation time and resource consumption.

[0037] Step S103: Based on the scanned surface point cloud set and the standard surface point cloud set, determine the surface correspondence relationship between each scanned surface point cloud and the standard surface point cloud, so as to determine the corresponding multiple surface transformation matrices based on the surface correspondence relationship.

[0038] Based on the scanned surface point cloud set and the standard surface point cloud set, the surface correspondence relationship of each scanned surface point cloud and the standard surface point cloud is determined. Specifically, this includes: obtaining multiple scanned surface point clouds in the scanned surface point cloud set and multiple standard surface point clouds in the standard surface point cloud set; extracting point cloud features for each scanned surface point cloud and each standard surface point cloud to determine the scanned surface features corresponding to each scanned surface point cloud and the standard surface features corresponding to each standard surface point cloud; and performing feature matching on each scanned surface point cloud and the standard surface point cloud based on each scanned surface feature and the standard surface feature to determine the specified standard surface point cloud corresponding to each scanned surface point cloud, thereby determining the surface correspondence relationship.

[0039] In one embodiment of this specification, feature extraction is performed on each scanned surface point cloud and each standard surface point cloud. The extracted features are used to describe the shape, geometric properties, and local structure of the point clouds. Commonly used point cloud features include local surface descriptors (such as FPFH, SHOT, PFH, etc.), curvature, normal vector distribution, point cloud density, etc. For each point cloud, the feature values ​​of all its points are calculated to form a feature vector or feature map. The extracted features are then used to match the scanned surface point cloud with the standard surface point cloud.

[0040] The feature vectors or feature maps of each point cloud are transformed into a common feature space for comparison. Similarity metrics such as Euclidean distance and cosine similarity are used to evaluate the similarity between the features of two point clouds. Matching algorithms (such as nearest neighbor search, FLANN, KD-tree, etc.) are used to find the most similar corresponding point cloud in the standard surface point cloud set for each scanned surface point cloud. Based on the feature matching results, for each scanned surface point cloud, the standard surface point cloud with the highest similarity is selected as its corresponding point cloud, thus establishing the surface correspondence relationship for each scanned surface point cloud. For example, if the point cloud data of the standard tube segment is split into six surfaces a, b, c, d, e, and f, and the point cloud data of the tube segment scan is split into four surfaces a', c', d', and f', the corresponding relationships are a' with a, c' with c, d' with d, and f' with f. Among them, a' and a are point clouds of the same type and size, c' and c are point clouds of the same type and size, d' and d are point clouds of the same type and size, and f' and f are point clouds of the same type and size. The transformation operation of the point cloud will be calculated based on these correspondences.

[0041] Based on the surface correspondence, multiple surface transformation matrices are determined, specifically including: determining a specified scanned surface point cloud and a corresponding specified standard surface point cloud through the surface correspondence; performing standardization processing on the specified scanned surface point cloud and the specified standard surface point cloud to determine the processed scanned surface point cloud and the processed standard surface point cloud corresponding to the specified scanned surface point cloud, and determining the standardization transformation matrix and the corresponding standardization transformation inverse matrix; determining the flipped scanned surface point cloud and the flipped transformation matrix based on the processed scanned surface point cloud and the processed standard surface point cloud; performing registration based on the flipped scanned surface point cloud and the processed standard surface point cloud using an iterative nearest-point algorithm to determine the registration transformation matrix; and determining multiple corresponding surface transformation matrices based on the standardization transformation matrix, the standardization transformation inverse matrix, the flipped transformation matrix, and the registration transformation matrix.

[0042] Based on the processed scanned surface point cloud and the processed standard surface point cloud, the flipped scanned surface point cloud and the flip transformation matrix are determined. Specifically, this includes: performing a flip operation on the processed scanned surface point cloud according to a preset flip method to determine multiple candidate flipped scanned surface point clouds, wherein the flip method includes rotating around the X-axis by a specified angle, rotating around the Y-axis by a specified angle, and rotating around the Z-axis by a specified angle; matching the multiple candidate flipped scanned surface point clouds and the processed scanned surface point cloud with the processed standard surface point cloud respectively to determine the flipped scanned surface point cloud; and determining the flip transformation matrix based on the flipped scanned surface point cloud.

[0043] In one embodiment of this specification, a specified scanned surface point cloud and a corresponding specified standard surface point cloud are determined through the surface correspondence relationship. Here, the specified scanned surface point cloud and the corresponding specified standard surface point cloud refer to two surfaces that have a surface correspondence relationship. Let pcd_std_major represent the specified standard surface point cloud and pcd_scan_major represent the specified scanned surface point cloud. The point cloud pcd_std_major is normalized to obtain the processed standard surface point cloud pcd_std_major_cent, and two transformation matrices std_cmtx and std_dcmtx. These two transformation matrices refer to the standard transformation matrices corresponding to the standard surface point cloud. The point cloud pcd_scan_major is normalized to obtain the processed scanned surface point cloud pcd_scan_major_cent, and two transformation matrices: the normalized transformation matrix scan_cmtx and the corresponding normalized inverse transformation matrix scan_dcmtx.

[0044] It should be noted that point cloud standardization can be achieved through the following steps: First, calculate the centroid of the point cloud. The centroid (also called the center point or centroid) is the average position of all points in the point cloud, providing a reference point for subsequent transformations. The centroid can be obtained by calculating the average coordinates of all points in the point cloud. Second, perform Principal Component Analysis (PCA) on the point cloud. PCA is a statistical method used to analyze patterns in data and extract the most important features (i.e., principal components). In point cloud processing, PCA is used to find the three principal directions of the point cloud data (i.e., the three orthogonal bases). These directions represent the three dimensions with the greatest variation in the data; these three directions are typically the "length," "width," and "height" of the point cloud data. Third, obtain the standardization transformation matrix and its inverse using the centroid and the three orthogonal bases obtained from PCA. A transformation matrix can be constructed using the centroid and the three orthogonal bases obtained from PCA. This transformation matrix transforms the original point cloud from the original coordinate system to a new coordinate system, where the origin of the new coordinate system is located at the centroid, and the coordinate axes are aligned with the three orthogonal bases obtained from PCA. This transformation process is called standardization or normalization. At the same time, it is also necessary to calculate the inverse matrix of this transformation so that the standardized point cloud can be transformed back to the original coordinate system when needed.

[0045] Flipping `pcd_scan_major_cent` 180° relative to the three coordinate axes yields four alternative schemes: the point cloud itself (`pcd_scan_major_cent`), and three alternative flipped scan surface point clouds corresponding to the point clouds rotated 180° around the X-axis, Y-axis, and Z-axis. These three alternative flipped scan surface point clouds and the processed scan surface point cloud are then matched with the processed standard surface point cloud. The scheme closest to the processed standard surface point cloud `pcd_std_major_cent` is identified as the flipped scan surface point cloud `pcd_scan_major_cent_flipped`. Two transformation matrices are generated using `pcd_scan_major_cent` and `pcd_scan_major_cent_flipped`: the flip transformation matrix `flip_cmtx` and its corresponding flip transformation inverse matrix `flip_dcmtx`. By using the Iterative Closest Point (ICP) algorithm,

[0046] Registering pcd_scan_major_cent_flipped to pcd_std_major_cent yields two transformation matrices: the registration transformation matrix reg_cmtx and the corresponding registration transformation inverse matrix reg_dcmtx.

[0047] Standardization eliminates differences in scale and position between the scanned surface point cloud and the standard surface point cloud, providing a more consistent basis for subsequent alignment and registration. Flipping ensures that the scanned surface point cloud is consistent with the target standard surface point cloud in a specific direction, reducing registration errors caused by inconsistencies in orientation. The ICP algorithm finds the optimal transformation matrix through iterative optimization, further improving registration accuracy and enabling the scanned surface point cloud to be as close as possible to the standard surface point cloud. Standardization and flipping increase the robustness of the algorithm, allowing it to handle point cloud data of different scales, positions, and orientations. By determining the correspondence between surfaces, fine-grained registration can be performed for each local surface region.

[0048] Step S104: The point cloud data of the tunnel segment scan is registered with the point cloud data of the standard tunnel segment through multiple surface transformation matrices to determine the registered point cloud data of the tunnel segment. Based on the registered point cloud data of the tunnel segment, the quality of the concrete tunnel segment to be inspected is carried out.

[0049] Using these multiple surface transformation matrices, the scanned point cloud data of the tunnel segment is registered with the point cloud data of the standard tunnel segment to determine the registered point cloud data of the tunnel segment. Specifically, this includes: determining the normalization transformation matrix, the inverse normalization transformation matrix, the flip transformation matrix, and the registration transformation matrix in the surface transformation matrix; using the normalization transformation matrix, performing point cloud normalization processing on the scanned point cloud data of the tunnel segment to determine the normalized scanned point cloud data of the tunnel segment; flipping the normalized scanned point cloud data of the tunnel segment according to the flip transformation matrix to determine the flipped scanned point cloud data of the tunnel segment; performing registration transformation on the flipped scanned point cloud data of the tunnel segment based on the registration transformation matrix to determine the registered scanned point cloud data of the tunnel segment; and transforming the registered scanned point cloud data of the tunnel segment using the inverse normalization transformation matrix to determine the registered point cloud data of the tunnel segment.

[0050] In one embodiment of this specification, the pcd_scan point cloud data of the tube segment is processed as follows: the pcd_scan point cloud is normalized using the normalization transformation matrix scan_cmtx to obtain normalized tube segment scan point cloud data pcd_scan_cent; the normalized tube segment scan point cloud data pcd_scan_cent is flipped using the flip transformation matrix flip_cmtx to obtain flipped tube segment scan point cloud data pcd_scan_cent_flipped; the pcd_scan_cent_flipped is transformed using the registration transformation matrix reg_cmtx to obtain registered tube segment scan point cloud data pcd_scan_cent_reg; and the inverse normalization transformation matrix std_dcmtx is used to transform it to the standard part point cloud position to obtain pcd_scan_reg tube segment registered point cloud data, which is the point cloud to be registered after being registered with the standard part point cloud. It should be noted that the registration process described above is the registration of each surface. The registration of the complete point cloud can be achieved by registering each surface. In addition, the pcd_scan point cloud in the above scheme refers to any surface point cloud in the tube segment scan point cloud data, and its corresponding multiple surface transformation matrices are the set of transformation matrices corresponding to this surface point cloud.

[0051] The application of the standardized transformation matrix eliminates the differences in scale and position between the scanned point cloud and the standard part point cloud, providing a unified basis for subsequent registration; the flip transformation matrix ensures the consistency between the scanned point cloud and the target standard part point cloud in a specific direction, reducing registration errors caused by inconsistencies in direction; the registration transformation matrix is ​​calculated through optimization methods such as the ICP algorithm, which can further reduce the deviation between the scanned point cloud and the standard part point cloud, achieving high-precision registration; by determining the surface correspondence and calculating multiple surface transformation matrices, fine registration of each local surface region can be achieved.

[0052] Based on the segment registration point cloud data, the quality inspection of the concrete segment to be inspected is carried out, specifically including: extracting multi-dimensional features from the segment registration point cloud data to determine the corresponding dimensional feature data and geometric feature data; using the dimensional feature data, performing segment size inspection on the concrete segment to be inspected to determine the corresponding dimensional deviation data; matching the geometric feature data with the standard geometric feature data corresponding to the pre-acquired segment standard component point cloud data to determine the corresponding shape consistency matching data of the concrete segment to be inspected; and determining the corresponding quality assessment information of the concrete segment to be inspected based on the dimensional deviation data and the shape consistency matching data, so as to carry out segment quality inspection on the concrete segment to be inspected.

[0053] In one embodiment of this specification, tools from point cloud processing libraries (such as PCL, Open3D, etc.) are used to extract key dimensional information, such as length, width, height, and diameter, from the registered point cloud data. This information can be obtained by fitting geometric shapes such as planes and cylinders and calculating their parameters. The extracted dimensional features are compared with the corresponding dimensions in the standard part's point cloud data to calculate the dimensional deviation. The dimensional deviation data can be expressed as the difference or percentage between the actual size and the standard size. Geometric features such as curvature and normal vectors of the point cloud are extracted, and the extracted geometric feature data is matched with the geometric features of the standard part's point cloud data. Feature matching algorithms (such as FLANN, KD-Tree, etc.) can be used to find the most similar set of feature points. The similarity or distance between the matched feature point sets is calculated as a measure of shape consistency, i.e., shape consistency matching data. Based on the dimensional deviation data and shape consistency matching data, quality assessment criteria are established. Thresholds can be set to determine whether the dimensional deviation and shape consistency are within acceptable ranges. Based on the assessment results, a quality assessment report is generated, including details of the dimensional deviation and a shape consistency score, to achieve quality inspection of the concrete pipe segment to be inspected. Multi-dimensional feature extraction can more comprehensively reflect the geometric characteristics of the tunnel segment, including size and shape details, reducing detection errors caused by insufficient extraction of a single feature. By matching precise size feature data and geometric feature data, fine detection of tunnel segment size and shape can be achieved, improving the accuracy and reliability of detection. Size feature data covers key dimensional information of the tunnel segment, such as length, width, height, and diameter, and can comprehensively assess whether the tunnel segment's size meets the standards. Geometric feature data focuses on the shape and surface characteristics of the tunnel segment, such as curvature and normal vector, and can reflect the overall shape consistency and surface quality of the tunnel segment. Combining size and geometric feature data can achieve a comprehensive assessment of tunnel segment quality.

[0054] By extracting surface point clouds from the scanned point cloud data and standard point cloud data of the tunnel segment separately, the actual surface shape of the tunnel segment can be more accurately focused on, rather than noise or irrelevant parts in the entire point cloud data. This reduces registration errors caused by missing data, as surface point cloud extraction typically retains key shape information while removing unnecessary background or noise. Determining the surface correspondence between each scanned surface point cloud and the standard surface point cloud, and calculating multiple surface transformation matrices based on this correspondence, is a more refined and accurate registration method. Compared to traditional global registration methods, it can consider the local differences between different surfaces of the tunnel segment, thus achieving more accurate local registration and reducing registration errors caused by incomplete data or scanning angle limitations. Quality inspection based on the registered point cloud data of the tunnel segment ensures the accuracy and reliability of the inspection results. Since the registration process has minimized errors caused by missing data and inaccurate registration, the subsequently extracted dimensional and geometric feature data will be closer to the actual situation, improving the accuracy and reliability of quality inspection.

[0055] This specification also provides an embodiment of a concrete tunnel segment quality inspection device, such as... Figure 2 As shown, the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described method.

[0056] This specification also provides a non-volatile computer storage medium storing computer-executable instructions configured to execute the above-described method.

[0057] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0058] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0059] The devices, media, and methods provided in the embodiments of this specification are one-to-one correspondences. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0060] Those skilled in the art will understand that embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0065] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0066] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0067] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0068] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for inspecting the quality of concrete tunnel segments, characterized in that, The method includes: Acquire the point cloud data of the concrete segment to be inspected and the point cloud data of the standard component of the concrete segment to be inspected. Surface point cloud extraction is performed on the scanned point cloud data of the tube segment and the point cloud data of the standard tube segment to determine the scanned surface point cloud set and the standard surface point cloud set. The scanned surface point cloud set includes multiple scanned surface point clouds, and the standard surface point cloud set includes multiple standard surface point clouds. Based on the scanned surface point cloud set and the standard surface point cloud set, determine the surface correspondence relationship between each scanned surface point cloud and the standard surface point cloud, and determine multiple corresponding surface transformation matrices based on the surface correspondence relationship. The point cloud data of the tunnel segment scan is registered with the point cloud data of the tunnel segment standard component by using the multiple surface transformation matrices to determine the tunnel segment registration point cloud data, so as to perform tunnel segment quality inspection on the concrete tunnel segment to be inspected based on the tunnel segment registration point cloud data. Surface point cloud extraction is performed on the scanned point cloud data of the tunnel segment and the point cloud data of the standard tunnel segment to determine the scanned surface point cloud set and the standard surface point cloud set, specifically including: Each scanned point cloud in the tube segment scanned point cloud data and each standard point cloud in the tube segment standard component point cloud data are analyzed to determine the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud. The scanned point cloud feature information includes scanned point distance information and scanned normal vector information, and the standard point cloud feature information includes standard point distance information and standard normal vector information. Based on the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud, the scanned surface point cloud set and the standard surface point cloud set are determined. Based on the scanned point cloud feature information of each scanned point cloud and the standard point cloud feature information of each standard point cloud, a scanned surface point cloud set and a standard surface point cloud set are determined, specifically including: Based on the scanning point distance information of each of the scanned point clouds and the standard point distance information of each of the standard point clouds, noise points are removed from the tube segment scanned point cloud data and the tube segment standard part point cloud data respectively to determine the first tube segment scanned point cloud data and the first tube segment standard part point cloud data. Based on the scanning normal vector information of each of the scanning point clouds, determine multiple target scanning surface point clouds in the first tube segment scanning point cloud data, and obtain the number of scanning points in each of the target scanning surface point clouds; Using the standard normal vector information of each standard point cloud, multiple target standard surface point clouds in the point cloud data of the first pipe segment standard component are determined, and the number of standard points in each target standard surface point cloud is obtained. Based on a preset first point count threshold and the number of scan points in each target scanned surface point cloud, the plurality of target scanned surface point clouds are filtered to determine the plurality of scanned surface point clouds, wherein the number of scan points in each scanned surface point cloud is not less than the first point count threshold. Based on the second point count threshold and the number of standard points in each of the target standard surface point clouds, the plurality of target standard surface point clouds are filtered to determine the plurality of standard surface point clouds, wherein the number of standard points in each of the standard surface point clouds is not less than the second point count threshold.

2. The method for inspecting the quality of concrete tunnel segments according to claim 1, characterized in that, Based on the scanned surface point cloud set and the standard surface point cloud set, determine the surface correspondence between each scanned surface point cloud and the standard surface point cloud, specifically including: Obtain multiple scanned surface point clouds from the scanned surface point cloud set and the standard surface point cloud set includes multiple standard surface point clouds; Point cloud features are extracted for each of the scanned surface point clouds and each of the standard surface point clouds to determine the scanned surface features corresponding to each of the scanned surface point clouds and the standard surface features corresponding to each of the standard surface point clouds. Based on each scanned surface feature and the standard surface feature, feature matching is performed on each scanned surface point cloud and the standard surface point cloud to determine the specified standard surface point cloud corresponding to each scanned surface point cloud, thereby determining the surface correspondence relationship.

3. The method for inspecting the quality of concrete tunnel segments according to claim 1, characterized in that, Based on the surface correspondence, determine the corresponding multiple surface transformation matrices, specifically including: The specified scanned surface point cloud and the corresponding specified standard surface point cloud are determined by the surface correspondence relationship. The specified scanned surface point cloud and the specified standard surface point cloud are standardized respectively to determine the processed scanned surface point cloud corresponding to the specified scanned surface point cloud and the processed standard surface point cloud corresponding to the specified standard surface point cloud, and the standardized transformation matrix and the corresponding standardized transformation inverse matrix are determined. Based on the processed scanned surface point cloud and the processed standard surface point cloud, determine the flipped scanned surface point cloud and the flipping transformation matrix; The registration transformation matrix is ​​determined by registering the flipped scanned surface point cloud and the processed standard surface point cloud using the iterative nearest point algorithm. Based on the standardized transformation matrix, the standardized transformation inverse matrix, the flip transformation matrix, and the registration transformation matrix, a plurality of corresponding surface transformation matrices are determined.

4. The method for inspecting the quality of concrete tunnel segments according to claim 3, characterized in that, Based on the processed scanned surface point cloud and the processed standard surface point cloud, the flipped scanned surface point cloud and the flipping transformation matrix are determined, specifically including: According to a preset flipping method, the processed scanned surface point cloud is flipped to determine multiple candidate flipped scanned surface point clouds. The flipping method includes rotating around the X-axis by a specified angle, rotating around the Y-axis by a specified angle, and rotating around the Z-axis by a specified angle. The multiple candidate flipped scanned surface point clouds and the processed scanned surface point cloud are matched with the processed standard surface point cloud to determine the flipped scanned surface point cloud. The flip transformation matrix is ​​determined based on the flipped scan surface point cloud.

5. The method for inspecting the quality of concrete tunnel segments according to claim 1, characterized in that, The segment scanning point cloud data and the segment standard part point cloud data are registered using the multiple surface transformation matrices to determine the segment registration point cloud data, specifically including: Determine the normalized transformation matrix, the normalized inverse transformation matrix, the flipped transformation matrix, and the registration transformation matrix in the surface transformation matrix; The point cloud data of the tube segment scan is standardized by using the standardized transformation matrix to determine the standardized tube segment scan point cloud data. The standardized tube segment scan point cloud data is flipped according to the flip transformation matrix to determine the flipped tube segment scan point cloud data. Based on the registration transformation matrix, the flipped tube segment scan point cloud data is registered and transformed to determine the registered tube segment scan point cloud data; The standardized transformation inverse matrix is ​​used to transform the registered segment scan point cloud data to determine the segment registration point cloud data.

6. The method for inspecting the quality of concrete tunnel segments according to claim 1, characterized in that, Based on the segment registration point cloud data, the quality inspection of the concrete segments to be inspected is carried out, specifically including: Multi-dimensional feature extraction is performed on the registration point cloud data of the tunnel segments to determine the corresponding size feature data and geometric feature data; Using the dimensional feature data, the dimensions of the concrete segment to be inspected are measured to determine the dimensional deviation data corresponding to the concrete segment to be inspected. The geometric feature data is matched with the geometric feature data of the standard parts corresponding to the pre-acquired point cloud data of the standard parts of the pipe segment to determine the shape consistency matching data of the concrete pipe segment to be inspected. Based on the dimensional deviation data and the shape consistency matching data, the quality assessment information corresponding to the concrete segment to be inspected is determined in order to perform segment quality inspection on the concrete segment to be inspected.

7. A quality inspection device for concrete tunnel segments, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-6.

8. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are configured to perform the method as described in any one of claims 1-6.

Images