Title :
Systematic static shadow detection
Author :
Yao, Jian ; Zhang, Zhongfei
Author_Institution :
Dept. of Comput. Sci., New York State Univ., Binghamton, NY, USA
Abstract :
A systematic static shadow detection algorithm for color images is presented in this paper. The image is modeled by an undirected graph and the shadow detection is achieved through maximizing the graph probability using the EM algorithm. Further analysis shows the connection between our model and the relaxation-labeling (RL) model. Experiments clearly indicate that our method is superior to a state-of-the-art shadow detection algorithm.
Keywords :
graph theory; image colour analysis; object detection; probability; color images; graph probability; relaxation-labeling model; static shadow detection; undirected graph; Color; Computer science; Detection algorithms; Information geometry; Labeling; Lighting; Object detection; Optimization methods; Pixel; Random variables;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334044
