CN-121981992-A - Defect detection method and related device for outer surface of wall pipe

CN121981992ACN 121981992 ACN121981992 ACN 121981992ACN-121981992-A

Abstract

The invention belongs to the technical field of detection, and relates to a defect detection method and a related device for the outer surface of a wall pipe. The method comprises the steps of obtaining three-dimensional point cloud data of the outer surface of a wall pipe to be detected, constructing a reference curved surface model describing the geometric form of the outer surface of the wall pipe based on the three-dimensional point cloud data, calculating normal deviation vectors from each measuring point in the three-dimensional point cloud data to the reference curved surface model, identifying areas on the outer surface of the wall pipe, wherein the areas exceed a preset threshold value, based on the normal deviation vectors of the measuring points, and judging the areas as defect areas. The invention realizes the accurate detection of the surface defects of the wall pipe.

Inventors

XU XIONGFEI
HU BOYAO
CHENG JUN
QIN CHENGPENG
LI DONGJIANG
WANG ZHIQIANG
WANG QIANG
LI LIANG
WANG FUGUI
CHEN ZHENG
ZHAO LUN

Assignees

西安热工研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (10)

1. A method for detecting defects on an outer surface of a wall pipe, comprising the steps of: Acquiring three-dimensional point cloud data of the outer surface of a wall pipe to be detected; constructing a reference curved surface model describing the geometric form of the outer surface of the wall pipe based on the three-dimensional point cloud data; calculating normal deviation vectors from each measuring point in the three-dimensional point cloud data to the reference curved surface model; and identifying a region on the outer surface of the wall pipe, which exceeds a preset threshold value, based on the normal deviation vector of each measuring point, and judging the region as a defect region.
2. The method for detecting defects on an outer surface of a wall pipe according to claim 1, wherein after the three-dimensional point cloud data of the outer surface of the wall pipe to be detected is obtained, the method further comprises: performing space grid division on the three-dimensional point cloud data; Based on the divided grids, processing the three-dimensional point cloud data by adopting a three-dimensional outlier detection algorithm, and eliminating outlier noise points in the three-dimensional point cloud data.
3. The method for detecting defects on an outer surface of a wall pipe according to claim 1, wherein constructing a reference surface model describing the geometry of the outer surface of the wall pipe based on the three-dimensional point cloud data comprises: topological dividing the three-dimensional point cloud data along the axial direction and the circumferential direction of the wall pipe to form a plurality of dividing units; for each dividing unit, a local surface fitting algorithm is adopted to acquire surface parameters representing the geometric form of the dividing unit; and based on the curved surface parameters of the plurality of dividing units, splicing and integrating the curved surface parameters through a smoothness constraint condition to form the reference curved surface model.
4. The method for detecting defects on an outer surface of a wall pipe according to claim 1, wherein the calculating a normal deviation vector from each measurement point in the three-dimensional point cloud data to the reference surface model comprises: Determining a dividing unit and a corresponding local reference curved surface to which each measuring point belongs for each measuring point; Calculating the shortest distance from the measuring point to the local reference curved surface, and determining an included angle between the shortest distance direction and a normal vector of the local reference curved surface at a projection point; When the included angle is smaller than or equal to a preset angle tolerance, the shortest distance is used as the normal deviation quantity; and when the included angle is larger than a preset angle tolerance, marking the measuring point as a point to be rechecked, and recalculating the normal deviation vector of the point to be rechecked based on the reference curved surfaces of the adjacent dividing units.
5. The method for detecting defects on an outer surface of a wall pipe according to claim 1, wherein after the area is determined as a defective area, further comprising: Extracting defect characteristics of the defect area, wherein the defect characteristics at least comprise depth, area, volume, contour shape and extension length along the axial direction and the circumferential direction of the wall pipe of the defect; And matching the defect characteristics with a preset defect characteristic database, and classifying defect types based on the matching result, wherein the defect types comprise at least one of pits, bulges, cracks and corrosion pits.
6. The method for detecting defects on the outer surface of a wall pipe according to claim 5, wherein the method for constructing the preset defect characteristic database comprises the following steps: Obtaining standard defect sample data with defect type labels; Extracting defect characteristics of the defect sample for each defect sample in the standard defect sample data to form a standard defect characteristic vector; and correlating the standard defect feature vector with the defect type corresponding to the standard defect feature vector to obtain a defect feature database.
7. The method for detecting defects on an outer surface of a wall pipe according to claim 6, wherein the classifying the types of defects based on the matching result comprises: Calculating the similarity between the defect characteristic vector of the defect area and each standard defect characteristic vector in the defect characteristic database; and determining the defect type of the defect area according to the defect type associated with the standard defect characteristic vector with the highest similarity.
8. A defect detection system for an outer surface of a wall tube, comprising: the point cloud data acquisition module is used for acquiring three-dimensional point cloud data of the outer surface of the wall pipe to be detected; the reference modeling module is used for constructing a reference curved surface model describing the geometric form of the outer surface of the wall pipe based on the three-dimensional point cloud data; the deviation amount calculation module is used for calculating the normal deviation amount from each measuring point in the three-dimensional point cloud data to the reference curved surface model; And the defect identification module is used for identifying the area on the outer surface of the wall pipe, which exceeds a preset threshold value, based on the normal deviation vector of each measuring point, and judging the area as a defect area.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1-7.

Description

Defect detection method and related device for outer surface of wall pipe Technical Field The invention belongs to the technical field of detection, and relates to a defect detection method and a related device for the outer surface of a wall pipe. Background The geometric integrity of the outer surface of the wall tubing industrial component is critical during manufacture and use. The defects of pits, bulges, scratches or corrosion and the like on the surface not only directly affect the appearance and the structural strength of the product, but also become the source of fatigue crack initiation, and form potential threat to the safety and the service life of equipment. Traditional detection methods such as manual visual or contact measurement have inherent limitations such as low efficiency, strong subjectivity, easy omission and difficult quantification. In recent years, along with the progress of three-dimensional scanning technology, high-density surface point cloud data can be rapidly obtained through means such as laser scanning or structured light, and a data base is provided for realizing automatic detection. However, how to accurately and automatically identify the tiny local geometric anomalies from a huge amount of discrete point clouds is still a technical difficulty in practical application. The existing method is often directly dependent on local curvature or roughness analysis of point cloud, has insufficient robustness on the overall geometric form (such as cylindrical surface bending, taper change and the like) of a wall pipe, is easy to misjudge the normal macroscopic curved surface form as a defect, and causes that the detection precision and reliability are difficult to meet the requirements of high industrial standards. Disclosure of Invention The invention provides a defect detection method and a related device for the outer surface of a wall pipe, which are used for solving the problems in the prior art and realizing the accurate detection of the surface defects of the wall pipe. In order to achieve the purpose, the invention is realized by adopting the following technical scheme: In a first aspect, the present invention provides a defect detection method for an outer surface of a wall pipe, comprising the steps of: Acquiring three-dimensional point cloud data of the outer surface of a wall pipe to be detected; constructing a reference curved surface model describing the geometric form of the outer surface of the wall pipe based on the three-dimensional point cloud data; calculating normal deviation vectors from each measuring point in the three-dimensional point cloud data to the reference curved surface model; and identifying a region on the outer surface of the wall pipe, which exceeds a preset threshold value, based on the normal deviation vector of each measuring point, and judging the region as a defect region. Preferably, after the obtaining the three-dimensional point cloud data of the outer surface of the wall pipe to be detected, the method further includes: performing space grid division on the three-dimensional point cloud data; Based on the divided grids, processing the three-dimensional point cloud data by adopting a three-dimensional outlier detection algorithm, and eliminating outlier noise points in the three-dimensional point cloud data. Preferably, the constructing a reference curved surface model describing the geometric form of the outer surface of the wall pipe based on the three-dimensional point cloud data includes: topological dividing the three-dimensional point cloud data along the axial direction and the circumferential direction of the wall pipe to form a plurality of dividing units; for each dividing unit, a local surface fitting algorithm is adopted to acquire surface parameters representing the geometric form of the dividing unit; and based on the curved surface parameters of the plurality of dividing units, splicing and integrating the curved surface parameters through a smoothness constraint condition to form the reference curved surface model. Preferably, the calculating a normal deviation vector from each measurement point in the three-dimensional point cloud data to the reference curved surface model includes: Determining a dividing unit and a corresponding local reference curved surface to which each measuring point belongs for each measuring point; Calculating the shortest distance from the measuring point to the local reference curved surface, and determining an included angle between the shortest distance direction and a normal vector of the local reference curved surface at a projection point; When the included angle is smaller than or equal to a preset angle tolerance, the shortest distance is used as the normal deviation quantity; and when the included angle is larger than a preset angle tolerance, marking the measuring point as a point to be checked, and recalculating the normal deviation vector based on the reference curved surfaces of th