DocumentCode :
3699049
Title :
Color cast detection for surveillance video using additive and subtractive color system
Author :
Ping Yang;Li Chen;Jing Tian
Author_Institution :
School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan University of Science and Technology, Wuhan, China, 430065
fYear :
2015
Firstpage :
1
Lastpage :
5
Abstract :
To tackle color cast detection challenge, a new approach is proposed in this paper to detect color cast and estimate the extent of color cast using additive and subtractive color system. First, the proposed approach calculates bright and dark information and extreme intensity ratio, and then determines the type of bright and dark information according to extreme intensity ratio. Finally, the proposed approach determines color system type and cast variation channel, and estimates the extent of color cast using extreme intensity ratio and difference intensity. Experimental results show that the proposed approach is reliable and effective, as well as consistent to human´s subjective evaluation.
Keywords :
"Image color analysis","Additives","Histograms","Channel estimation","Neural networks","Estimation","Surveillance"
Publisher :
ieee
Conference_Titel :
Signal Processing, Communications and Computing (ICSPCC), 2015 IEEE International Conference on
Print_ISBN :
978-1-4799-8918-8
Type :
conf
DOI :
10.1109/ICSPCC.2015.7338967
Filename :
7338967
Link To Document :
بازگشت