Premium
Automatic visual inspection method for a coated surface having an orange peel effect
Author(s) -
Yimit Adiljan,
Itou Akiyoshi,
Matsui Youichi,
Akashi Takuya
Publication year - 2019
Publication title -
ieej transactions on electrical and electronic engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 30
eISSN - 1931-4981
pISSN - 1931-4973
DOI - 10.1002/tee.22824
Subject(s) - visual inspection , production line , artificial intelligence , computer science , computer vision , orange (colour) , line (geometry) , engineering , optics , mathematics , mechanical engineering , physics , geometry
In many production processes, defect detection has played an important role in preventing the outflow of defective products. As such, there are still many production lines that rely on an inspector's visual observation. In this study, using fluorescent tubes as an irradiation light source, we propose a simple and effective visual inspection method for detecting convex defect on painted surfaces having an orange peel effect. The proposed method detects and distinguishes painted surface defects in an area illuminated by fluorescent tubes, using the combination of local segmented results and gradient information computed with different differential operations. The inspection device for detecting the defect can be easily installed in the production line without changing the production line, and can be constructed at a low cost without using additional adaptations. The performance of the proposed method was evaluated with the data of real products captured with an inspection device installed on the production line. Experimental results show that this method can provide a better detection in the inspection region with a low false‐detection rate. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.