Design and Development of an Intelligent Vision System for Wiper Arm Defect Detection with Novel Tunnel Lighting System

Ng, Yu Ming (2021) Design and Development of an Intelligent Vision System for Wiper Arm Defect Detection with Novel Tunnel Lighting System. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Ng Yu Ming_Full text.pdf Restricted to Registered users only Download (2MB)

Abstract

Quality control is an essential process of ensuring that a product has fulfilled the quality requirements and customer satisfactions. In car wiper industry, the quality of the wipers is a key aspect to stay competitive in market and have good customer collaboration. The defect inspection of the wipers can be the solution to maintain the quality of the wipers. However, the defect inspection system has the limitations due to the undetected defects and high computational load. There are many defect inspection methods to detect the defects of the wipers. The aim of this research is to improve the accuracy of the defect inspection system for surface defects compared to the previous defect inspection system. The autoencoder model and the YOLOv3 model are applied to detect the surface defects of the wipers with the novel lighting system. The overall accuracy of the autoencoder model and the YOLOv3 model was 77.87 percent and 91.80 percent respectively.