Method for detecting the notch deviation based on corner point constraint

2018 International Conference on Construction, Aviation and Environmental Engineering

生态环境科学;生物科学

Xiao, Zhitao^1^2 ; Wei, Quansheng^1^2 ; Geng, Lei^1^2

School of Electronics and Information Engineering, Tianjin Polytechnic University, Tianjin, Tianjin

300387, China^1

Tianjin Key Laboratory of Optoelectronic Detection Technology and System, Tianjin Polytechnic University, Tianjin, Tianjin

300387, China^2

关键词: Affine transformations;    Constraint methods;    Corner point;    Detection methods;    Harris corner detection;    Neighbourhood;    Triangle areas;    Triangle constraint;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/233/3/032014/pdf
DOI  :  10.1088/1755-1315/233/3/032014

来源: IOP

【 摘 要 】

The notch detection plays a critical role in influencing the quality control of cut-piece. It is difficult to measure the deviation of notch due to the deformation of cut-piece, which in turn affects the subsequent sewing process and the quality of cut-piece. In this paper, a notch detection method based on corner point constraint is proposed to solve this problem. Firstly, all the corner points are extracted from the CAD by using the improved Harris corner detection algorithm. The triangle constraint and the intersection constraint method are used to eliminate extra corner points that are not in the triangle area, and then the notch is obtained. Secondly, the cut-piece and the CAD are superimposed by affine transformation, and the approximate notch area on the cut-piece is obtained in the neighbourhood of the notch on the CAD. Finally, the improved Harris corner detection algorithm is used to extract all the corner points of the notch area on the cut-piece, and then the notch is extracted after removing the extra corner points. The experimental results have shown a good consistency with the expected performance which confirms the effectiveness of the proposed method.