Title :
A Novel Facial Image Retrieval Algorithm Based on PLT-PCA
Author :
Han, Qi ; Song, Chen ; Zhang, Meihui ; Niu, Xiamu
Author_Institution :
Harbin Inst. of Technol., Harbin
Abstract :
Face Image Retrieval is an important application of Content-Based Image Retrieval. A more practical model of face image retrieval system is analyzed in this paper. Moreover a suitable algorithm based on power law transformation(PLT) is introduced. The feature space is spanned by the training set enlarged from the input query image rather than images in the retrieving dataset. Experiments and comparison with projection based method show the feasibility and efficiency of the proposed algorithm.
Keywords :
content-based retrieval; face recognition; image retrieval; principal component analysis; PLT-PCA; content-based image retrieval; facial image retrieval; power law transformation; principal component analysis; projection based method; Content based retrieval; Face recognition; Hidden Markov models; Image databases; Image retrieval; Indexes; Information retrieval; Principal component analysis; Signal processing algorithms; Spatial databases; PCA; face image retrieval; power law transformation;
Conference_Titel :
Intelligent Information Hiding and Multimedia Signal Processing, 2008. IIHMSP '08 International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-0-7695-3278-3
DOI :
10.1109/IIH-MSP.2008.217
