Title :
Computational model of selective attention for machine vision based on adapted entropy
Author :
Tian, Yan-tao ; Lian, Tao ; Yu, Da-Chuan ; Xiao, Jie-Wei
Author_Institution :
Coll. of Commun. & Eng., Jilin Univ., Changchun, China
Abstract :
The selective attention for machine vision can reduce the complexity of calculation. This paper adapted entropy as a measurement to the salience of interested object region and proposed a new computational model of attention selection. Analysis proved that this method simulated the selective attention of human being effectively and was easy for engineering realization. Experiments showed the feasibility and efficiency of the calculation model.
Keywords :
computer vision; entropy; object recognition; adapted entropy; computational model; machine vision; object region; salience map; selective attention; Analytical models; Computational modeling; Computer vision; Educational institutions; Entropy; Humans; Image processing; Layout; Machine vision; Pixel; Entropy; Machine vision; Salience map; Selective attention;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527896
