Identification of PCB faults using image processing

The aim of this paper is to develop a more accurate PCB inspection system. In this method captured colored images are natural images i.e. without any precustomized environment and hence the every time the capture images varies with lighting conditions, the distance and inclination and these issues are to be handled effectively. Most of the work in this regard just detects the location of the fault but merely little work has been carried out to identify the type of the PCB faults automatically. This paper attempts to detect two types of PCB faults effectively.