11th International Conference on Image Analysis and Processing (ICIAP'01)
Image Analysis of Crosswalk
Palermo, Italy
September 26-September 28
ISBN: 0-7695-1183-X
Abstract: This paper proposes a method for image analysis of a crosswalk and a traffic light. The method provides information not only about a length of a crosswalk, but also about a colour of traffic light. A length of a crosswalk is estimated by the projective geometry using white lines painted on a road at a crosswalk. Furthermore, a state of a traffic light, that is, a colour of green (walk signal) or red (stop signal), is detected by searching a green (or red) traffic light using affine moment invariants. In order to evaluate the performance, experimental results of estimating a length and of detecting a traffic light are presented.
Citation:
Tadayoshi Shioyama, Haiyuan Wu, Yuna Nishibe, Naoki Nakamura, Suguru Kitawaki, "Image Analysis of Crosswalk," iciap, pp.0168, 11th International Conference on Image Analysis and Processing (ICIAP'01), 2001