Retrieval and recognition of faces using content-based image retrieval (CBIR) and feature combination method
Ningthoujam Sunita Devi, Kattamanchi Hemachandran · 2016
Face recognition is one of the most successful applications of image analysis and understanding and has gained much attention in last decades. In this paper, we purpose a simple and fast hybrid face recognition system based on CBIR and SVM. The Gabor wavelets (GW), Wavelet Transformation (WT), and principal component analysis (PCA) are used as feature extraction methods to generate a feature vector. The Euclidean Distance as a similarity measurement used to retrieve similar images and fed into the SVM is for recognition. The experimental results demonstrate that the proposed work on face recognition system outperforms the existing solution in terms of the system running time and recognition accuracy. The experimental results also indicate that the fusion of PCA, GW, and WT features as a feature vector performs reasonably well with 99.9% recognition accuracy of the proposed system and can be suitable for real life applications.