Leader ImViA : Olivier Aubreton

ABSTRACT :

Recently, the global industry is facing a fourth revolution, called Industry 4.0. It refers to the connectivity and integration of smart and autonomous systems to collect and analyze data and digitize the organization’s operations. Indeed, by adding smart cameras to production lines, we are seeing improved defect detection quality at high speeds and reduced costs. Moreover, given the restrictions imposed on human labor by the COVID-19 law, such as social distancing in the factory, these benefits are even more essential to keep production lines functional. In order to address this issue, our goal in this project is to create a digital twin of the production line based on image analysis and Artificial Intelligence (AI). The aim is to link the AI expertise of the DISC department and the know-how in non-conventional imaging of the IMVIA laboratory. This twin would allow the generation of realistic synthetic images. These images model the different acquisition systems set up on the production line. They will then be used to train an AI for defect detection. In a second step, the functional production line will be used to feed the database. Another goal of the project is to answer the lack of industrial image database.

Partners: FEMTO-ST DISC

Beginning: 05/01/2022

End: 04/11/2025

WebSite: https://gradschool.eiphi.ubfc.fr/?p=3693