Research Article
A Novel Full-Angle Detection Method for Bolt Loosening Based on Color Segmentation
@INPROCEEDINGS{10.4108/eai.17-11-2023.2342751, author={Jingjie Kang and Lijun Zhang and Yuandong Sun and Xiaoyu Yang and Ruolan Wang and Tianhao Zhao}, title={A Novel Full-Angle Detection Method for Bolt Loosening Based on Color Segmentation}, proceedings={Proceedings of the First International Conference on Science, Engineering and Technology Practices for Sustainable Development, ICSETPSD 2023, 17th-18th November 2023, Coimbatore, Tamilnadu, India}, publisher={EAI}, proceedings_a={ICSETPSD}, year={2024}, month={1}, keywords={bolt looseness detection full-angle detection color segmentation nonlinear stretching vector calculation}, doi={10.4108/eai.17-11-2023.2342751} }- Jingjie Kang
Lijun Zhang
Yuandong Sun
Xiaoyu Yang
Ruolan Wang
Tianhao Zhao
Year: 2024
A Novel Full-Angle Detection Method for Bolt Loosening Based on Color Segmentation
ICSETPSD
EAI
DOI: 10.4108/eai.17-11-2023.2342751
Abstract
In response to the existing problem of the inability to perform full-angle detection of bolt loosening based on anti-loosening lines segmentation, this paper develops a method based on color segmentation and directional vector calculation. Initially, the anti-loosening line image is segmented using an optimal threshold segmentation method for the R component and a non-linear stretching method for the a component under the transformation of Lab and RGB color spaces. Subsequently, morphological operations are performed on the image using an open operation. Then, the minimum bounding rectangle of the anti-loosening line connected domain area is determined through an angle progressive minimum encapsulation method, and its directional vector is established. Consequently, the full-angle of bolt loosening is calculated based on the four-quadrant arctangent function and specific adjustments. Finally, an experimental scheme is designed to validate the feasibility and accuracy of the proposed algorithm. The experimental results show that the detection algorithm can achieve a 0 to 360-degree loosening angle detection of bolts, with a maximum absolute error of 0.6345 degrees and a maximum relative error of 0.75%. The accuracy of the algorithm can meet the needs of engineering practice and exhibits significant potential for application.