Research on Detection Method of Sheet Surface Defects Based on Machine Vision

Abstract

The paper is based on machine vision technology to study the detection of sheet surface defects, designs the hardware system and software processing flow of online detection, introduces processing methods such as image preprocessing, image segmentation and target extraction, and processes the defect images accordingly. Use C# software to design the human-computer interaction interface and on-line debugging and on-line detection for the online detection of sheet defects. The experimental results show that the detection method is feasible, the false detection rate is low, and it can be well applied to the online detection of wood automatic production process.

Defect identification method of macrostructure crack based on position information feature

“…The fabrication of macrostructure samples requires complex processes such as cutting, corrosion and polishing. The defect detection of macrostructure cracks requires professionals to mark their position with the naked eye and evaluate the defect grade according to the national standard [3][4][5]. Factors such as working hours affect the result of manual division of defects, which is easy to lead to unstable division results.…”