Design of a Gaussian filter based on Particle swarm optimization for automatic visual inspection system

This paper proposes a development of image processing technique for inspecting the printed circuit board (PCB) of hard disk drive (HDD), which is an important part. Normally, PCB inspection is carried out by using high-resolution line scan camera. The visual inspection system of PCB is usually corrupted by image processing noise. Measuring system is significantly affected by noise because the size of the PCB is very small. To overcome this problem, this paper proposes an effective filter which uses Particle swarm optimization to design the optimal standard deviation parameter in the Gaussian filter. Experimental results demonstrate the effectiveness of the proposed filtering technique.