A Machine Vision Based Automatic Optical Inspection System for Detecting Defects of PCBA

Abstract

The process of manufacturing electronic printed circuit board assembly (PCBA) is a common process in technology companies. In particular, error checking of a PCBA with full components is a difficult job for human inspectors due to the high concentration and persistence. To solve this problem, many solutions have been studied, in which image processing technology has emerged as the best solution with advantages such as fast processing speed and high accuracy. However, the cost of these systems is often quite high for small and medium size companies. This paper presents a solution using image processing technology to check for missing components on complete PCBAs. Accordingly, a suitable algorithm is proposed for low-cost cameras, which allows building a simpler and more economical inspection process for both types of solder joints: through hole and surface mounting device (SMD). Consuming time to test a PCBA with nearly 50 components is less than 1 s. A cheap and easy-to-use Automatic Optical Inspection (AOI) system had been developed and used to test the actual PCBAs at the company.