DocumentCode
3392675
Title
Comparative studies of visual stimuli features affect the stimulus recognize
Author
Liu Jun-ling ; Liu Jun-yi ; Guan Rong-qiang ; An Xiaofeng
Author_Institution
Coll. of Comput. Sci. & Technol., Jilin Univ., Changchun, China
fYear
2011
fDate
19-22 Aug. 2011
Firstpage
917
Lastpage
920
Abstract
The experimental methods of psychology provides a basis of computer vision simulation,because it can reveal the law of human visual information processing. The shape and color are important features of stimulations, which are very crucial for the identification of stimulations. This study is based on the experimental psychology and BP neural networks, use experimental methods of reaction time measurements and feature extraction methods of the BP neural network, by the experiment of identifying three groups of geometric images of the same color in different shapes ,the same shape in different colors and shape and color are both different, and attempt to explore the relationships of affect of the stimulus recognition between shape features and color features of stimulations in the visual information process. Comparing and analyzing the experimental results getting from the experimental psychology methods, we can find that: in the image feature recognition, recognizing the shape feature is faster than the color feature.And the same experimental results obtained by BP neural network simulation.
Keywords
backpropagation; computer vision; feature extraction; image colour analysis; neural nets; visual perception; BP neural network; color feature; computer vision simulation; experimental psychology; feature extraction; geometric image; human visual information processing; image feature recognition; reaction time measurement; shape feature; stimulation identification; stimulus recognition; visual stimuli feature; Accuracy; Feature extraction; Humans; Image color analysis; Image recognition; Psychology; Shape; BP neural network; E-PRIME; reaction time(RT); stimuli features; stimulus recognize;
fLanguage
English
Publisher
ieee
Conference_Titel
Mechatronic Science, Electric Engineering and Computer (MEC), 2011 International Conference on
Conference_Location
Jilin
Print_ISBN
978-1-61284-719-1
Type
conf
DOI
10.1109/MEC.2011.6025614
Filename
6025614
Link To Document