Cited 8 time in
Cited 10 time in
Illumination-Robust Face Recognition System Based on Differential Components
- Illumination-Robust Face Recognition System Based on Differential Components
- Lee, SH[Lee, Sang-Heon]; Kim, DJ[Kim, Dong-Ju]; Cho, JH[Cho, Jin-Ho]
- DGIST Authors
- Lee, SH[Lee, Sang-Heon]; Kim, DJ[Kim, Dong-Ju]
- Issue Date
- IEEE Transactions on Consumer Electronics, 58(3), 963-970
- Article Type
- Biometrics; Consumer Applications; D2D-PCA; Differential Component; Face Images; Face Recognition; Face Recognition Systems; Facial Feature; Illumination Conditions; Illumination Effect; Illuminationvariation; Matching Score; Performance Degradation; Performance Evaluation; Principal Component Analysis; Recognition Accuracy; Subimages; Two-Dimensional Principal Component Analysis; Yale Face Database
- Illumination variation generally causes performance degradation of face recognition systems under real-life environments. Therefore, we propose an illuminationrobust face recognition system using a fusion approach based on efficient facial feature called differential two-dimensional principal component analysis (D2D-PCA) for consumer applications. In the proposed method, face images are divided into two sub-images to minimize illumination effects, and D2D-PCA is separately applied to each sub-images. The individual matching scores obtained from two sub-images are then integrated using a weighted-summation operation, and the fused-score is utilized to classify the unknown user. Performance evaluation of the proposed system was performed using an extended Yale face database B which consists of 2,414 face images for 38 subjects representing 64 illumination conditions under the frontal pose. Experimental results show that the proposed fusion approach enhanced recognition accuracy by 22.02% compared to that of 2DPCA, and we confirmed the effectiveness of the proposed face recognition system under illumination-variant environments © 2011 IEEE.
- Institute of Electrical and Electronics Engineers Inc.
- Related Researcher
There are no files associated with this item.
- Convergence Research Center for Future Automotive Technology1. Journal Articles
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.