An Automated Ceramic Tiles Inspection System using Image Processing Technique with Cloud Management System

Abstract

Globally, the Ceramic tiles technical industrial sector is a comparatively new but very huge industry that has taken significant gain of the robust development of automation in the past few years. This industry these days hold significant ratio of consumer need. For industries, the measurement and perfection of tiles analysis is a challenging task. Ceramic tiles process through different stages of production with technologies except the final process of inspecting the defective tiles, which, done manually by human vision in many manufacturing industries. Detecting the defects manually takes a lot of human effort in enormous industries, which costs highly and is time-consuming. With human error factor involved in inspection process, there are evident cases where defected pieces gone unnoticed into the market and it creates a negative impact of any industry. The system proposed will detect visual defects such as cracks, pinholes, spots on the surface of the tiles with acute accuracy. It will also detect irregularities in patterns printed on the tiles. The proposed system provides a state of the art solution to measurement and defect analysis in the tiles using image-processing techniques. The outcome of this project is to detect defects and pattern mismatch in the tiles, then indicating any defect signaling red or green light using raspberry pie and mentioning that defect ratio with an inspection report as decision parameters. The work described here comprises of two modules ¾  an automated Ceramic measurement and defect analysis system (AMTDS) using image processing techniques and cloud management. The AMTDS is an effective and low cost technique, which minimizes the issue of accurate measurement and defect analysis of Ceramic tiles with the accuracy (87.5%) and precision (80.3%).