Open AccessComparative Study of Image Processing and Transfer Learning Techniques for an Automated PCB Fault Detection System
Author(s) -
Muskaan Rajput
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.35775
Subject(s) - printed circuit board , computer science , fault detection and isolation , image processing , transfer of learning , process (computing) , fault (geology) , artificial intelligence , subtraction , quality (philosophy) , background subtraction , image (mathematics) , automated optical inspection , reliability engineering , embedded system , real time computing , engineering , pixel , mathematics , philosophy , arithmetic , epistemology , seismology , actuator , geology , operating system
Printed circuit board (PCB) is one of the most crucial components in most electronic devices. PCBs are manufactured in large quantities, and therefore, maintaining the quality of such large numbers of PCBs is important. An automated inspection system can help in the aspect of quality maintenance. Such a system is able to overcome the limitations of manual inspection for a large number of PCBs. It provides fast detection of defects and hence can prove to be an asset in the manufacturing process. This project aims to achieve fault detection of bare PCBs through two different methods; the traditional algorithmic approach using image processing, which makes use of image subtraction method, and a transfer learning approach, which involves the pre-trained CNN VGG16 model. A comparative study of both the methods is done.