Research On Weld Pool Over Melted Feature Extraction With Contour Extraction Based On Chan-Vese With Fuzzy C-Means

2018 INTERNATIONAL CONFERENCE ON ROBOTS & INTELLIGENT SYSTEM (ICRIS 2018)(2018)

引用 1|浏览109
暂无评分
摘要
Aiming at the weld pool over melted phenomenon when the welding speed is too slow, the weld pool image segmentation based on Chan-Vese (CO model with Fuzzy C-Means (FCM) is proposed. Based on the contour extraction of the weld pool image under different welding speeds, the Roundness characteristic parameter is proposed to describe the phenomenon of weld pool over melted. The results show that the FCM-CV algorithm eliminates the problem that the CV model is sensitive to the initial contour, and the FCM-CV algorithm can extract the contours of the weld pool under different technological conditions with good adaptability and stability. Roundness well reflects the weld pool characteristics under different welding speeds, which can well describe the phenomenon of the weld pool over melted, when the Roundness is larger than 0.9.
更多
查看译文
关键词
Feature Extraction, Active Contour, Fuzzy Clustering, Weld Pool Characteristic
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要