Mask R-CNN Based System for Accurate Detection and Analysis of Chip Damage on Car Paint

This thesis presents the development of a Mask R-CNN based system for detecting chip damages on car paintings and analyzing their dimensions. The project was carried out at Toyota, a global company that requires a uniform and efficient system for detecting chip damages without the need for specialized technicians to travel around the world. The system utilizes advanced computer vision algorithms and deep learning techniques to detect and analyze chip damages on car paintings with high accuracy. It can detect chip damages and provide measurements of their dimensions with a significantly higher speed and accuracy respect to human inspection. Overall, the system developed in this project has the potential to significantly improve the efficiency of chip damage detection and analysis in the automotive industry, benefiting companies like Toyota that operate on a global scale.