Approaches to the development of a printed circuit board defect detection system based on AOI technology

T. S. Khodataeva, N. V. Kashirin, A. I. Averina, A. E. Guryanov

Mari State University

Abstract: Some modern approaches to detecting defects in printed circuit boards based on automatic optical inspection are considered in order to design their own control system. The importance of the control process is growing in connection with the tightening of the requirements imposed by modern production processes. At the enterprises of mass production of electronics, attempts are being made to achieve high quality of all parts, assemblies and finished products. The optical inspection system is one of the most important tools for automating the visual inspection of printed circuits. In addition to ensuring cost efficiency and product quality control, an automated control system can also collect statistical information to provide feedback to the production process. The review considers algorithms and methods for automated optical control of the conductive pattern on the surface of printed circuit boards in order to find the optimal method for detecting defects.

Keywords: automatic optical inspection, image processing, convolutional neural networks

DOI: 10.15514/ISPRAS-2023-35(4)-5