Author : Prachi P. Londe and S. A. Chavan

The importance of the Printed Circuit Board inspection process has been magnified by requirements of the modern manufacturing environment. In electronics mass production manufacturing facilities, an attempt is often to achieve 100% quality assurance. In this work Machine Vision PCB Inspection System is applied at the first step of manufacturing. In this system a PCB inspection system is proposed and the inspection algorithm mainly focuses on the defect detection and defect classification of the defects. Defect classification is essential to the identification of the defect sources. The purpose of the system is to provide the automatic defect detection of PCB and relieve the human inspectors from the tedious task of finding the defects in PCB which may lead to electric failure. We first compare a standard PCB inspection image with a PCB image to be inspected. The MATLAB tool is used to detect the defects and to classify the defects. With the solutions designed and implemented in this thesis the algorithm designed in the proposed system is able to detect and classify all the known 14 types of defects successfully with greater accuracy. The algorithm makes use of image subtraction method for defect detection and kNN classification algorithm for the classification of the defects. This thesis will present and analyze the performance of the proposed inspection algorithm. The experiment will measure the accuracy of the system.

Keywords: PCB- Printed circuit board, PCB Defects detection, Classification

Article published in International Journal of Current  Engineering  and Technology, Vol.4,No.3 (June- 2014)