Detection method of micro and macro scale defects in silicon carbide castings

• Jinyang Wang, Jingjing Chen, and Haiping Shen
• received 12/28/2023; accepted 04/23/2024; posted 04/23/2024; Doc. ID 517305
• Abstract: Aiming at challenges such as low efficiency, high missing rate, difficulty inidentifying contour defects , and difficulty in extracting tiny defects, a defect detection methodfor extracting micro and macro scale defects was proposed in this paper. After preprocessingthe image, contour detection is performed to identify the contours. Subsequently, a contourcomplementation algorithm is employed to complement the unclosed contours. Finally, thedetection of micro scale defects is conducted based on the grayscale variation of the center of themicro scale efects. The experimental results show that compared with the traditional method, theproposed algorithm can accurately detect the bubble defects of different scales in silicon carbidecastings, and can identify the complex defects better.