The complexity of electronic circuits is increasing due to novel products such as smartphones, smart TVs, smartwatches, drones, and robots. Furthermore, the competition between the electronic companies is fiercer than ever. In this sense, product quality is a key point in the company’s business plan. Being the Printed Circuit Board (PCB) the main and most sensitive part of an electronic product, ensuring its quality is fundamental to the production process. Visual Inspection is one of the most important tools to ensure PCB quality in a manufacturing line. Considering the advances in computing systems, the Automated Optical Inspection (AOI) machines are replacing human operators in the visual inspection process. However, AOI machines usually still need highly trained operators, with knowledge of image processing, so there still has room for more automated processes using Artificial Intelligence, particularly considering the advances with Neural Networks. This is the subject of this paper, where it is presented a survey about the traditional techniques and neural networks techniques able to identify components and anomalies in various PCB types. At the end of the paper, the works are compared highlighting the pros and cons of each methodology when applied to a real production line.
CITATION STYLE
Andrade, M. A., Pepe, P. C. F., Ximenes, L. R., & Arthur, R. (2022). A Survey on Automatic Inspection for Printed Circuit Board Analysis. In Smart Innovation, Systems and Technologies (Vol. 295 SIST, pp. 423–431). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-08545-1_40
Mendeley helps you to discover research relevant for your work.