Research on Weld Identification and Defect Localization Based on an Improved Watershed Algorithm
The quality of weld joints is a pivotal factor influencing the strength and structural reliability for mechanical parts. Difficulties in identification caused by weld joint adhesion and defects, such as missing weld joints, this paper designs an efficient welded joint detection system, which utilize...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10829943/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The quality of weld joints is a pivotal factor influencing the strength and structural reliability for mechanical parts. Difficulties in identification caused by weld joint adhesion and defects, such as missing weld joints, this paper designs an efficient welded joint detection system, which utilizes Otsu threshold segmentation and morphological processing methods to achieve the initial segmentation of the welded joint region. Furthermore, the contour of the weld joint is extracted with greater accuracy by random incremental algorithm, which contributes to accelerating the following the detection speed only considering the weld joint region. Subsequently, watershed algorithm based on distance transformation is adopted to segment each weld joint precisely. Considering defect localization, the edge features of the missing weld region are identified and separated using the difference calculation method. The results demonstrate that the system could accurately segment adhesive weld joints and identify missing weld joint locations with the variation of the workpiece positions and angles, fully satisfying the real-time detection requirement during the weld quality identification. |
|---|---|
| ISSN: | 2169-3536 |