CN-120976188-B - Welding steel structure quality detection method

CN120976188BCN 120976188 BCN120976188 BCN 120976188BCN-120976188-B

Abstract

The application relates to the technical field of image processing and provides a welding steel structure quality detection method which comprises the steps of collecting welding line images of a welding steel structure to be detected, dividing welding line areas, identifying each group of adjacent jump points according to gray values of adjacent pixel points in the same data row in the welding line areas, dividing a gas hole area to be confirmed according to difference of gray values of the pixel points between the adjacent jump points in all the adjacent groups in the same data row, determining annular area contrast and gas hole confidence of the gas hole area to be confirmed, and obtaining quality detection results of the welding steel structure to be detected according to the gas hole confidence of all the gas hole areas to be confirmed in the welding line images of all different welding lines of the welding steel structure to be detected. The application aims to improve the accuracy of welding air hole defect detection.

Inventors

REN LIANG
ZHANG HENG
MENG XINXIN
LI LIANG
SHEN DEZHOU
MA LEI
ZHANG ZHONGHUA
YUAN WEIQIANG
ZHENG CHAO
LIANG PENGFEI
LU XIUYU

Assignees

青岛亚维迪精密金属制造有限公司

Dates

Publication Date: 20260508
Application Date: 20250825

Claims (7)

1. The method for detecting the quality of the welded steel structure is characterized by comprising the following steps of: acquiring a welding seam image of a steel structure to be detected, and dividing a welding seam region in the welding seam image; identifying each group of adjacent jumping points according to the gray values of adjacent pixel points in the same data row in the welding line region, and dividing the region of the air hole to be confirmed according to the gray value difference of the pixel points between the adjacent jumping points in all the same data row; determining the contrast of the annular region of the air hole region to be confirmed according to the gray value difference of the inside and the periphery of the air hole region to be confirmed, and calculating the air hole confidence coefficient of the air hole region to be confirmed by combining the circularity of the air hole region to be confirmed; Acquiring a quality detection result of the welded steel structure to be detected according to the air hole confidence degrees of all air hole areas to be confirmed in the weld joint images of all different weld joints of the welded steel structure to be detected; the method for acquiring the annular region contrast of the air hole region to be confirmed comprises the following steps: The minimum circumcircle of the air hole area to be confirmed is marked as a first outer circle of the air hole area to be confirmed; Constructing a first outer circular ring of the air hole area to be confirmed according to the first outer circular ring of the air hole area to be confirmed; the average value of gray values of all pixel points contained in the first outer circular ring of the air hole area to be confirmed is recorded as a first gray value of the air hole area to be confirmed; The average value of gray values of all pixel points contained in the first outer circle of the air hole area to be confirmed is recorded as a second gray value of the air hole area to be confirmed; determining the annular area contrast of the air hole area to be confirmed according to the difference between the first gray value and the second gray value of the air hole area to be confirmed; the construction method of the first outer circular ring comprises the following steps: Taking the center of a first outside circle of the air hole area as the center of a circle, taking the radius length of the first outside circle of the air hole area plus the length of a first preset number of pixel points as the radius, constructing a circle, and marking the circle as a second outside circle of the air hole area to be confirmed; the area where the second outer circle of the air hole area to be confirmed is not overlapped with the first outer circle is marked as a first outer circular ring of the air hole area to be confirmed; The method for calculating the contrast of the annular region comprises the following steps: and marking the difference value between the first gray level and the second gray level of the air hole region to be confirmed as the first difference value of the air hole region to be confirmed, and marking the ratio of the first difference value to the second gray level of the air hole region to be confirmed as the annular region contrast of the air hole region to be confirmed.
2. The method for detecting the quality of a welded steel structure according to claim 1, wherein the dividing the weld region in the weld image comprises the following specific steps: respectively carrying out graying, denoising and edge detection treatment on the weld joint image to obtain a weld joint edge image; and marking the area which is divided by the two edges with the largest pixel number in the welding seam edge image as a welding seam area, and dividing the welding seam area in the welding seam image according to the pixel point positions in the welding seam area.
3. The method for detecting the quality of the welded steel structure according to claim 1, wherein the method for identifying the adjacent trip points is as follows: and calculating the gray value difference value of adjacent pixel points in the same data row in the welding line region, and marking the adjacent pixel points with the gray value difference value larger than a preset jump threshold value as a group of adjacent jump points.
4. The method for detecting the quality of the welded steel structure according to claim 1, wherein the method for dividing the air hole area to be confirmed is as follows: The average value of gray values of all pixel points between two adjacent groups of adjacent jump points in the same data line is recorded as the regional gray average value of a region formed by all pixel points between the two adjacent groups of adjacent jump points; and carrying out self-adaptive threshold segmentation on the gray average value of all the areas, obtaining a segmentation threshold value of the gray average value of the areas, and marking the areas formed by the pixel points corresponding to the gray average value of the areas smaller than the segmentation threshold value as the areas with air holes to be confirmed.
5. The method for detecting the quality of the welded steel structure according to claim 1, wherein the calculating the air hole confidence of the air hole region to be confirmed by combining the circularity of the air hole region to be confirmed comprises the following specific steps: Calculating the circularity of the air hole area to be confirmed; and (3) marking the product of the contrast ratio and the circularity of the annular region of the air hole region to be confirmed as the air hole confidence of the air hole region to be confirmed.
6. The method for detecting the quality of the welded steel structure according to claim 5, wherein the specific calculation method of the circularity is as follows: The number of the pixel points contained in the air hole area to be confirmed is calculated The product of the air hole area to be confirmed is recorded as a first product of the air hole area to be confirmed, and the ratio of the first product of the air hole area to be confirmed to the number of pixels contained at the edge of the air hole area to be confirmed is recorded as the circularity of the air hole area to be confirmed.
7. The method for detecting the quality of the welded steel structure according to claim 1, wherein the step of obtaining the quality detection result of the welded steel structure to be detected according to the air hole confidence level of all the air hole areas to be confirmed in the weld images of all the different welds of the welded steel structure to be detected comprises the following specific steps: Marking the air hole region to be confirmed with the air hole confidence coefficient larger than a preset air hole defect judging threshold value as an air hole defect region; When the total number of the identified air hole defect areas in the weld image of all the different weld joints of the to-be-detected welded steel structure is 0, judging that the to-be-detected welded steel structure has excellent quality; When the total number of the identified air hole defect areas in the weld image of all the different weld joints of the to-be-detected welded steel structure is larger than 0 and smaller than a preset number threshold value, judging that the to-be-detected welded steel structure has good quality; And when the total number of the identified air hole defect areas in the weld image of all the different welds of the welded steel structure to be detected is greater than or equal to a preset number threshold value, judging that the quality of the welded steel structure to be detected is unqualified.

Description

Welding steel structure quality detection method Technical Field The application relates to the technical field of image processing, in particular to a welding steel structure quality detection method. Background Welding is a main mode of steel structure connection, the quality of welding seams directly determines the bearing capacity and the anti-seismic performance of the whole structure, in order to ensure the safety of the welded steel structure, timely find and repair defects, prolong the service life of the structure, reduce the maintenance cost of the whole life cycle, and need to detect the quality of the welded steel structure of the building engineering in real time, and timely identify quality problems such as air holes, slag inclusion, unfused fusion and the like. Generally, when the quality of welded steel structures is detected, the characteristics of the air hole problem in welding can be extracted according to texture information in an image of a welding position, so that quality detection is realized. However, the partial characteristics of the air hole defects have similarity with other welding problems, so that other defects are easily misjudged as air hole defects, and the recognition accuracy of the air hole defect problems in welding quality detection is insufficient. Disclosure of Invention The application provides a welding steel structure quality detection method, which aims to solve the problems that partial characteristics of air hole defects are similar to other welding problems, so that the air hole defect detection is misjudged, and the recognition accuracy of the air hole defect problems in the welding quality detection is insufficient, and the adopted technical scheme is as follows: one embodiment of the application provides a welded steel structure quality detection method, which comprises the following steps: acquiring a welding seam image of a steel structure to be detected, and dividing a welding seam region in the welding seam image; identifying each group of adjacent jumping points according to the gray values of adjacent pixel points in the same data row in the welding line region, and dividing the region of the air hole to be confirmed according to the gray value difference of the pixel points between the adjacent jumping points in all the same data row; determining the contrast of the annular region of the air hole region to be confirmed according to the gray value difference of the inside and the periphery of the air hole region to be confirmed, and calculating the air hole confidence coefficient of the air hole region to be confirmed by combining the circularity of the air hole region to be confirmed; And acquiring a quality detection result of the welded steel structure to be detected according to the air hole confidence degrees of all the air hole areas to be confirmed in the weld joint images of all the different weld joints of the welded steel structure to be detected. Further, the specific method for dividing the weld joint area in the weld joint image comprises the following steps: respectively carrying out graying, denoising and edge detection treatment on the weld joint image to obtain a weld joint edge image; and marking the area which is divided by the two edges with the largest pixel number in the welding seam edge image as a welding seam area, and dividing the welding seam area in the welding seam image according to the pixel point positions in the welding seam area. Further, the method for identifying the adjacent hopping points comprises the following steps: and calculating the gray value difference value of adjacent pixel points in the same data row in the welding line region, and marking the adjacent pixel points with the gray value difference value larger than a preset jump threshold value as a group of adjacent jump points. Further, the method for dividing the air hole area to be confirmed comprises the following steps: The average value of gray values of all pixel points between two adjacent groups of adjacent jump points in the same data line is recorded as the regional gray average value of a region formed by all pixel points between the two adjacent groups of adjacent jump points; and carrying out self-adaptive threshold segmentation on the gray average value of all the areas, obtaining a segmentation threshold value of the gray average value of the areas, and marking the areas formed by the pixel points corresponding to the gray average value of the areas smaller than the segmentation threshold value as the areas with air holes to be confirmed. Further, the method for obtaining the annular region contrast of the air hole region to be confirmed comprises the following steps: The minimum circumcircle of the air hole area to be confirmed is marked as a first outer circle of the air hole area to be confirmed; Constructing a first outer circular ring of the air hole area to be confirmed according to the first outer circular ring of the air hole ar