Detection Algorithm of Porosity Defect on Surface of Micro-precision Glass Encapsulated Electrical Connectors

Abstract

A miniature precision glass encapsulated electrical connectors introduced by glass powder and metal wires through a special complicated process. Aiming at the porosity defects on the surface, a defect detection algorithm propose based on threshold segmentation and feature extraction. Pre-operation, global threshold segmentation processing and feature extraction (based on area, circularity aspect ratio, compactness, and contour length) are preformed to detect the defects. Experimental results show that the algorithm can accurately identify porosities defects.

Defect Detection of Micro-Precision Glass Insulated Terminals

“…The performance of the glass terminal has a huge impact on the operation of the equipment. If there is no strict inspection for defects before using glass terminals, it will cause huge safety hazards to the reliability of high-precision electronic equipment, bringing about serious losses and consequences [1].…”

“…Due to the limitation of production level and detection methods, some produced glass terminals have defects such as missing blocks, pores, and cracks (Figure 1). The difficulties in defect detection are mainly three points [2]: (1) The complex imaging background of the defects contains a variety of interference noise; (2) The shape, size, and location of defects are diverse; (3) Due to the different locations, sizes and shapes of missing blocks or pore defects, various defects will show greater differences. Therefore, this paper proposes to use deep learning technology to detect missing blocks [3].…”