Back to Search Start Over

The Study of Spatial Frequency Channels for Human Visual System.

Authors :
Xu, Xiangyang
Chen, Qiao
Xu, Ruixin
Source :
International Journal of Pattern Recognition & Artificial Intelligence. Jun2019, Vol. 33 Issue 6, pN.PAG-N.PAG. 17p.
Publication Year :
2019

Abstract

Similar to auditory perception of sound system, color perception of the human visual system also presents a multi-frequency channel property. In order to study the multi-frequency channel mechanism of how the human visual system processes color information, the paper proposed a psychophysical experiment to measure the contrast sensitivities based on 17 color samples of 16 spatial frequencies on CIELAB opponent color space. Correlation analysis was carried out on the psychophysical experiment data, and the results show obvious linear correlations of observations for different spatial frequencies of different observers, which indicates that a linear model can be used to model how human visual system processes spatial frequency information. The results of solving the model based on the experiment data of color samples show that 9 spatial frequency tuning curves can exist in human visual system with each lightness, R–G and Y–B color channel and each channel can be represented by 3 tuning curves, which reflect the "center-around" form of the human visual receptive field. It is concluded that there are 9 spatial frequency channels in human vision system. The low frequency tuning curve of a narrow-frequency bandwidth shows the characteristics of lower level receptive field for human vision system, the medium frequency tuning curve shows a low pass property of the change of medium frequent colors and the high frequency tuning curve of a width-frequency bandwidth, which has a feedback effect on the low and medium frequency channels and shows the characteristics of higher level receptive field for human vision system, which represents the discrimination of details. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
02180014
Volume :
33
Issue :
6
Database :
Academic Search Index
Journal :
International Journal of Pattern Recognition & Artificial Intelligence
Publication Type :
Academic Journal
Accession number :
135980017
Full Text :
https://doi.org/10.1142/S0218001419550073