Printed Circuit Boards (PCBs) have become an inevitable component in all electronic devices, including the gadgets that we use in our day-to-day life. To maintain the quality of a product it is necessary to make sure the PCB is void of defects. To identify whether PCB has defects or not, it has to undergo tests in various stages. In this paper, we propose a reference-based model using Wasserstein distance metric to detect short faults in PCBs. The study includes the effect of Gaussian and Salt-Pepper Noise have on fault detection, which usually occurs in the images procured by digital cameras.