Machine Vision for Excess Gluing Inspection in Spindle Motor Assembly

Quality inspection of gluing quality in inner threads of bearing sleeve of harddisk-drive is an important step in spindle motor assembly. This paper presents an automated inspection system based on machine vision for quality control of gluing quality. At present, the inspection process is performed by human inspectors. However, they have low throughput and cannot identify all defects. To increase the quality and productivity, an automated visual inspection system is needed. The approach is based on detecting reflected light from the defects under oblique lighting. The defects generate highlights along the threads in the image of the inner sleeve. To reduce spurious reflections from thread roofs, input image of threads is enhanced by a Gabor filtering and the threads are located by a geometric primitive fitting. By properly detecting those highlight spots within regions between adjacent thread roofs, the defect is accurately identified. Experimental result showed a good performance of our proposed algorithm.