Final Year Project

Project Title

Automatic Scratch Detection on Vehicle Assembly Line

Description

Precise detection of surface flaws, especially scratches and dents, is a difficult task for human vision. Our capstone design project is on the development of a low-cost system that uses vision algorithms to accurately identify scratches on a vehicle assembly line. The system can account for defects with random shapes, low contrast or that are obscured by the natural texture of the surface, whereas a robotic arm allows for movement of the camera system. A multi-imaging setup allows for predicting the exact source of the scratch. This work is being done in partnership with a local car manufacturing company and will help them add automated tools to their assembly line. This is a joint project with Muhammad Ashir Wahid and Syed Muhammad Hasan Naqvi.