A Novel Illumination Compensation Method Using Wavelet Packet Transformation

Zhen Hua Zhao; Xiao Hong Hao

doi:10.4028/www.scientific.net/AMR.317-319.897

Paper Titles

An Improved Pedestrian Detection Method Based on Adaboost and Performance Analysis
p.877

Lane Markers Detection Based on Consecutive Threshold Segmentation
p.881

Study on a Method for Measuring the Attitude of Protable Three-Coordinate Equipment Based on Monocular Vision
p.886

Research on Underwater Target Tracking Based on Contour Detection
p.890

A Novel Illumination Compensation Method Using Wavelet Packet Transformation
p.897

Handwritten Digit Recognition Based-on Bernoulli Mixtures
p.901

Fixed Viewpoint Imaging and Correction Model of Liquid Column Manometer
p.905

Applications of Visible and near-Infrared Hyperspectral Imaging for Non-Destructive Detection of the Agricultural Products
p.909

The Finite Ridgelet Transform for Defeat Detection of Quill
p.915

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 317-319A Novel Illumination Compensation Method Using...

A Novel Illumination Compensation Method Using Wavelet Packet Transformation

Abstract:

A novel illumination compensate method is proposed in this paper to improve recognition performance. A modified lighting model called Lambertin which includes additive noise and multiplicative noise are presented firstly. Then, additive noise is removed by using wavelet packet transformation. Next, the processed image is transformed into logarithm domain and the multiplicative noise, which has been named additive noise, is removed by means of the same above algorithm. Finally, a compensated face image is obtained. We examine the proposed method on Yale extended B database compared with other methods. Experimental results show that our algorithm improves by 3%~12% recognition rate. It can effectively adjust the facial images for varying illumination conditions and also improve the recognition performance and robustness.

You might also be interested in these eBooks

Equipment Manufacturing Technology and Automation

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 317-319)

Pages:

897-900

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.317-319.897

Citation:

Cite this paper

Online since:

August 2011

Authors:

Zhen Hua Zhao, Xiao Hong Hao

Keywords:

Additive Noise, Illumination Compensate, Logarithm Domain, Wavelet Packet (WP)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] ZHUANG L SH. Research on key algorithms for face recognition under complex illumination [D].Hefei: University of Science and Technology of China, 2006.(in Chinese)

Google Scholar

[2] ISH IYAMA R, SAKAMOTO S. Geodesic illumination basis: compensating for illumination variations in any pose for face recognition [C]. Proceeding of 16th Internat. Conf. Pattern recognition, 2002: 2972301.

DOI: 10.1109/icpr.2002.1047455

Google Scholar

[3] LEE S W, MOON S H, LEE S W. Face recognition under arbitrary illumination using illuminated exemplars [J]. Pattern Recognition, 2007, 40(5): 160521620.

DOI: 10.1016/j.patcog.2006.09.016

Google Scholar

[4] MOSES Y, ULLMAN S. Limitation of nonmodel-based recognition schemes[C] .Proceeding of ECCV292, Berlin: Springer Verlag, 1992: 8202828.

Google Scholar

[5] YAO J M, XU T F, NI G Q. Nonlinear target tracking method based on optimized wavelet features [J]. Opt. Precision Eng . , 2007, 15(3): 4282433. ( in Chinese)

Google Scholar

[6] SAVVIDESM, KUMARV. Illumination normalization using logarithm transform for face recognition[C] . Proceeding of IAPRAVBPA, 2010: 5492556.

Google Scholar

[7] XIE X D, LAMKM. Face recognition under varying illumination based on a 2D face shape model[J] . Pattern Recognition, 2009, 38(2) : 2212230.

DOI: 10.1016/s0031-3203(04)00275-4

Google Scholar

[8] ZHU JH , LIUB, SCHWARTZSC. General illumination correction and its application to face recognition[ C] .Proceeding of IEEE OCASSSP , 2003: 1332136.

Google Scholar