Edge extraction algorithm of metal parts based on binarization

Conference: MEMAT 2022 - 2nd International Conference on Mechanical Engineering, Intelligent Manufacturing and Automation Technology
01/07/2022 - 01/09/2022 at Guilin, China

Proceedings: MEMAT 2022

Pages: 4Language: englishTyp: PDF

Authors:
Gan,Yong (School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, Guangxi, China & Guangxi Key Laboratory of Manufacturing Systems and Advanced Manufacturing Technology, Guilin, China)
Rao, Chengjian; Zeng, Boqiao (School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, Guangxi, China)

Abstract:
Aiming at the problem that the traditional edge extraction algorithm is easily affected by illumination on the metal parts and there are false edges, a new edge detection algorithm based on binarization was proposed. First of all, the pixel value of the background and the metal parts will increase under the high light, but the pixel value of the metal parts will increase more than the background. In order to make the binarization effect obvious, the background color is black. Then, the pixel histogram of the image under high light is calculated, and the binarization threshold is obtained by using the bimodal method. Finally, the gradient changes of pixel values along X and Y axes are calculated to get the edges of metal parts and refine the edges. The results show that, compared with the traditional edge extraction algorithm, this algorithm is less affected by illumination, and the edge extraction is clear and the continuity is good.