Automatic detection system for verification and quality control: Application to water connector inspection
Visual inspection of finished products has always been one of the basic and most recognized applications of quality control in the industry field. This inspection remains largely a manual process conducted by operators, and thus faces considerable limitations that make it unreliable. Therefore, it is necessary to automatize this inspection for better efficiency. This work aims to explore the application of Machine Vision in a manufacturing system environment. The objective is to create and develop an automated inspection system used to sort the water connector product after the assembly process. The machine vision system solution allows to classify and sort the assembled finished water connector product as conforming product or non-conforming product. This work proposes two solutions to classify and identify the water connector product. The first inspection method is based on image segmentation while the second one is based on machine learning, achieving an accuracy of 96% and 87% respectively.