Date of Publication :25th July 2024

Abstract:Semiconductor technology is widely used in almost every field today be it communication, satellite technology, computing, energy, automotive, healthcare, e-commerce, retail, education, administration etc. A thin size of wafer is mainly used for computation and decision making process in electronic devices and apparatus, which is made up of semiconductor material also called printed circuit board or PCB. PCB consists of millions of logic circuits designed on its surface for different purposes. The major challenge in PCB manufacturing comes at the last stage where wafer has to go through the quality analysis and inspection process. This process had been on manual mode for a long time as the advent of automation came in industries, this step became automatic too. Yet the present systems used in defect detection are less efficient and time consuming. In this paper we propose to detect wafer defects using a CNN based deep learning model which will use the image of PCB and predict if the PCB is defected or non-defected in microseconds.

Reference :