In this paper we present a novel algorithm for face recognition using combination of wavelet transforms and principal component analysis (PCA). Face features are extracted using combination of Haar and Daubechies wavelet transform. Then obtained features are used for face recognition via PCA (eigenfaces). The experimental results show that the highest face recognition accuracy rate is obtained using the combination of Haar and Daubechies wavelet transforms for face features extraction. The proposed algorithm gives an effective performance of face recognition on noisy images and competes on the accuracy recognition with state-of-the-art algorithms. Some experiments were conducted on videos. The Viola-Jones method was used for face detection. The results were compared with «Associative neural networks» (ANN).
|Состояние||Опубликовано - 2016|
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition