Title :
Shape extraction of transparent object using genetic algorithm
Author :
Hata, Seiji ; Saitoh, Yoko ; Kumamura, Syoji ; Kaida, Ken´ichi
Author_Institution :
Fac. of Educ., Kagawa Univ., Takamatsu, Japan
Abstract :
There are many transparent objects to be measured in industrial fields. They are plastics, paste drops and liquid drops, etc. To have the high quality products, sometimes, it is very important to control the shape and the volume of these materials. But, to extract the shape of these objects by optical way is one of the difficult problems and not so many researches have been established in this field. In this paper, a shape extraction method of transparent objects using the combination of multi-stripe lighting and genetic algorithms is proposed. The method was experimentally applied to extract the shape of transparent paste drops and proved its efficiency
Keywords :
computer vision; feature extraction; genetic algorithms; lighting; stereo image processing; 3D shape models; computer vision; feature extract; genetic algorithm; multiple stripe lighting; shape extraction; transparent object; Deformable models; Educational products; Electronics industry; Genetic algorithms; Industrial electronics; Optical materials; Pattern analysis; Plastics; Shape control; Shape measurement;
Conference_Titel :
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
Conference_Location :
Vienna
Print_ISBN :
0-8186-7282-X
DOI :
10.1109/ICPR.1996.547652