TY - GEN
T1 - Illumination normalization for face recognition and uneven background correction using total variation based image models
AU - Chen, Terrence
AU - Yin, Wotao
AU - Zhou, Xiang Scan
AU - Comaniciu, Dorin
AU - Huang, Thomas S.
PY - 2005
Y1 - 2005
N2 - We present a new algorithm for illumination normalization and uneven background correction in images, utilizing the recently proposed TV+L 1 model: minimizing the total variation of the output cartoon while subject to an L1norm fidelity term. We give intuitive proofs of its main advantages, including the well-known edge preserving capability, minim.al signal distortion, and scale-dependent hut intensity-independent foreground extraction. We then propose a novel TV-based quotient image model (TVQI)for illumination normalization, an important preprocessing for face recognition under different lighting conditions. Using this model, we achieve 100% face recognition rate on Yale face database B if the reference images are under good lighting condition and 99.45% if not. These results, compared to the average 65% recognition rate of the quotient image model and the average 95% recognition rate of the more recent self quotient image model, show a clear improvement. In addition, this model requires no training data, no assumption on the light source, and no alignment between different images for illumination normalization. We also present the results of the related applications - uneven background correction for cDNA microarray films and digital microscope images. We believe the proposed works can serve important roles in the related fields.4.
AB - We present a new algorithm for illumination normalization and uneven background correction in images, utilizing the recently proposed TV+L 1 model: minimizing the total variation of the output cartoon while subject to an L1norm fidelity term. We give intuitive proofs of its main advantages, including the well-known edge preserving capability, minim.al signal distortion, and scale-dependent hut intensity-independent foreground extraction. We then propose a novel TV-based quotient image model (TVQI)for illumination normalization, an important preprocessing for face recognition under different lighting conditions. Using this model, we achieve 100% face recognition rate on Yale face database B if the reference images are under good lighting condition and 99.45% if not. These results, compared to the average 65% recognition rate of the quotient image model and the average 95% recognition rate of the more recent self quotient image model, show a clear improvement. In addition, this model requires no training data, no assumption on the light source, and no alignment between different images for illumination normalization. We also present the results of the related applications - uneven background correction for cDNA microarray films and digital microscope images. We believe the proposed works can serve important roles in the related fields.4.
UR - http://www.scopus.com/inward/record.url?scp=24644479273&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=24644479273&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2005.181
DO - 10.1109/CVPR.2005.181
M3 - Conference contribution
AN - SCOPUS:24644479273
SN - 0769523722
SN - 9780769523729
T3 - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
SP - 532
EP - 539
BT - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
PB - IEEE Computer Society
T2 - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
Y2 - 20 June 2005 through 25 June 2005
ER -