Quality control in porcelain industry based on computer vision techniques

Abstract

This paper presents a system based on computer vision techniques for quality monitoring the porcelain production flow. The quality monitored system is based on the robot-computer vision architecture and includes: (i) real-time high-speed processing of product images, and (ii) a global autonomous behaviour, context and task dependent self-learning that is adaptive to the work environment. We have investigated the use of integral Robot Vision (iRVision) technology. iRVision is a ready-to-use robotic vision package available for FANUC robots. The experimental evaluation shows that the inspection system that we developed can correctly identify if a product is defective or not. The proposed architecture will finally have a positive economic impact for the company by optimizing the production flow and reducing the production costs.