A Smart Monitoring System for Automatic Welding Defect Detection

This work introduces an intelligent system able to perform quality control assessment in an industrial production line. Deep learning techniques have been employed and proved successful in a real application for the inspection of welding defects on an assembly line of fuel injectors. Starting from state-of-the-art deep architectures and using the transfer learning technique, it has been possible to train a network with about 7 millions parameters using a reduced number of injectors images, obtaining an accuracy of 97,22%. The system has also been configured in order to exploit new data, collected during operation, to extend the existing dataset and to improve further its performance. The developed system showed that deep neural networks can successfully perform quality inspection tasks which are usually demanded to humans.