Title :
Illumination invariant algorithm for face image based on tricolor attenuation theory
Author :
Yang, Dawei ; He, Siyuan ; Tian, Jiandong ; Tang, Yandong
Author_Institution :
State Key Lab. of Robot., Shenyang Inst. of Autom., Shenyang, China
Abstract :
In the paper, a tricolor attenuation theory is introduced, a new algorithm which transform a color face image into a gray one which is illumination invariant based on the tricolor attenuation theory is proposed. The color face image is segmented into shadow part and non shadow part using tricolor attenuation theory, then computing the change rate between shadow region and non shadow region in each color channel, and structure a operator T with the changed rate. Finally, applying the operator T to the whole face image. The experiments demonstrate the validation of the algorithm.
Keywords :
face recognition; image colour analysis; image segmentation; color face image; gray face image; illumination invariant algorithm; image segmentation; tricolor attenuation theory; Attenuation; Face; Face recognition; Image color analysis; Image segmentation; Lighting; Training;
Conference_Titel :
Intelligent Control and Information Processing (ICICIP), 2012 Third International Conference on
Conference_Location :
Dalian
Print_ISBN :
978-1-4577-2144-1
DOI :
10.1109/ICICIP.2012.6391435