DocumentCode
305474
Title
Human sensitivity of color defects inspection
Author
Hata, Seiji ; Miyashita, Yoshihiro ; Hanafusa, Hideyuki
Author_Institution
Inf. & Comput. Sci. Lab., Kagawa Univ., Takamatsu, Japan
Volume
2
fYear
1996
fDate
5-10 Aug 1996
Firstpage
713
Abstract
Visual inspection is one of the important quality control functions of production lines. In this area, many automatic visual inspection machines have been introduced and proved their effectiveness. Today, visual inspection is going to be applied to the small and vague color defects inspection. In these cases, the measured color signal values do not correspond to human sensitivity. So, it is very important to evaluate the human sensitivity of color defects. There are three components to describe the color. They are brightness, hue and saturation. These parameters influence the human sensitivity to detect color defects. In this research, to develop the visual inspection system of color defects, human sensitivity changes according to the different color parameters are evaluated and in the color defects inspection machine, color signals are normalized by the evaluated human sensitivities. Through this normalization, the color visual inspection machine has the same ability with human inspectors
Keywords
automatic optical inspection; colour; colour vision; computer vision; quality control; automatic visual inspection machines; brightness; color defects inspection; color defects inspection machine; color visual inspection machine; hue; human sensitivity; measured color signal values; quality control functions; saturation; visual inspection; Brightness; Color; Computer science; Computer science education; Displays; Educational products; Humans; Inspection; Pixel; Production;
fLanguage
English
Publisher
ieee
Conference_Titel
Industrial Electronics, Control, and Instrumentation, 1996., Proceedings of the 1996 IEEE IECON 22nd International Conference on
Conference_Location
Taipei
Print_ISBN
0-7803-2775-6
Type
conf
DOI
10.1109/IECON.1996.565965
Filename
565965
Link To Document