[해외논문]
Recognition of multiple objects based on global image consistency between a scene hypothesis and the acquired image
Systems and computers in Japan = 電子情報通信 ,
v.32 no.2 ,
2001년, pp.19 - 31
Hashimoto, Manabu
(Industrial Electronics Systems Laboratory, Mitsubishi Electric Corporation, Amagasaki, 661-8661 Japan)
,
Kuroda, Shin'ichi
(Energy and Industrial Systems Center, Mitsubishi Electric Corporation, Hyogo, 652-8555 Japan)
,
Sumi, Kazuhiko
(Industrial Electronics Systems Laboratory, Mitsubishi Electric Corporation, Amagasaki, 661-8661 Japan)
,
Usami, Teruo
(Industrial Electronics Systems Laboratory, Mitsubishi Electric Corporation, Amagasaki, 661-8661 Japan)
,
Nakata, Shuji
(Department of Manufacturing Science, Osaka University, Suita, 565-0871 Japan)
In this paper, we describe an algorithm designed to recognize a large number of rectangular objects that have been placed near or in contact with one another. The conventional model-matching method, which identifies an object by evaluating the consistency between its acquired image and an object mod...
In this paper, we describe an algorithm designed to recognize a large number of rectangular objects that have been placed near or in contact with one another. The conventional model-matching method, which identifies an object by evaluating the consistency between its acquired image and an object model, frequently produces false results due to patterns and characters printed on the object. We propose to solve this problem by employing a new recognition method that maximizes the global consistency between a scene and an acquired image. This method is based on the fact that the correct interpretation of an entire scene results in higher overall consistency (e.g., global consistency) between the image generated from the interpretation and the actual image acquired. This recognition algorithm first generates multiple scene hypotheses of the locations and combinations of many candidate objects from a gray-scale edge image. Then, it selects the hypothesis that has the maximum global consistency with the acquired range images. This is a further development of an object recognition method that we proposed earlier and that is based on hypothesis verification utilizing model-matching with gray-scale edge images. Our method solves the problem of false recognition patterns that result from many objects being in contact with each other. The method handles the generation and verification of hypotheses as a problem of combining candidate objects, and solves this problem utilizing a genetic algorithm. In this study this proposed method successfully recognized objects 99.8% of the time in experiments using images of actual objects. This provides evidence that the method is practical for visual recognition systems in robots. © 2001 Scripta Technica, Syst Comp Jpn, 32(2): 19–31, 2001
In this paper, we describe an algorithm designed to recognize a large number of rectangular objects that have been placed near or in contact with one another. The conventional model-matching method, which identifies an object by evaluating the consistency between its acquired image and an object model, frequently produces false results due to patterns and characters printed on the object. We propose to solve this problem by employing a new recognition method that maximizes the global consistency between a scene and an acquired image. This method is based on the fact that the correct interpretation of an entire scene results in higher overall consistency (e.g., global consistency) between the image generated from the interpretation and the actual image acquired. This recognition algorithm first generates multiple scene hypotheses of the locations and combinations of many candidate objects from a gray-scale edge image. Then, it selects the hypothesis that has the maximum global consistency with the acquired range images. This is a further development of an object recognition method that we proposed earlier and that is based on hypothesis verification utilizing model-matching with gray-scale edge images. Our method solves the problem of false recognition patterns that result from many objects being in contact with each other. The method handles the generation and verification of hypotheses as a problem of combining candidate objects, and solves this problem utilizing a genetic algorithm. In this study this proposed method successfully recognized objects 99.8% of the time in experiments using images of actual objects. This provides evidence that the method is practical for visual recognition systems in robots. © 2001 Scripta Technica, Syst Comp Jpn, 32(2): 19–31, 2001
Keyword
참고문헌 (15)
Int J Robotics Res Bolles 5 3 1986 10.1177/027836498600500301
IEEE Trans PAMI Jain 10 783 1988 10.1109/34.9102
IEEE Trans PAMI Wheeler 17 252 1995 10.1109/34.368190
Computer Vision and Image Understanding Yi 69 87 1998 10.1006/cviu.1997.0597
Trans IEICE Fukada J76-D-II 1097 1996
Trans IEICE Shakunaga J78-D-II 1608 1995
Trans IEICE Kaku J80-D-II 1105 1997
Trans Third Image Sensing Symp Makita 291 1997
Trans IEICE Sotoyama J81-D-II 1584 1998
Trans First Image Sensing Symp Kimura 263 1995
Trans IEICE Tsunashima J79-D-II 795 1996
IEICE Trans Inf Syst Hashimoto E78-D 1552 1995
Trans IEICE Hashimoto J74-D-II 1419 1991
Hashimoto 1997 et al. A load recognition method based on a fusion of coarse range images and density edge information. Third Image Sensing Symp 1997, p 353-358.
Genetic algorithm in search, optimization, and machine learning Goldberg 1989 Genetic algorithm in search, optimization, and machine learning. Addison-Wesley; 1989.
더보기
AI-Helper
※ AI-Helper는 부적절한 답변을 할 수 있습니다.