An efficient PCA based pose and occlusion invariant face recognition system for video surveillance

摘要: In the proposed work introduced a pose-invariant face recognition process for videos so as todiminish calculation intricacy of traditional technique. The intended approach includes four phases, namely (i) preprocessing, (ii) object detection, (iii) feature extraction along with (iv) dimensionality reduction. initial phase is segmenting database video clips into frames where preprocessing done using an adaptive median filtering such that to eradicate noise. subsequent detecting preprocessed image through viola–jones With this technique, mouth, left eye, face, nose, right eye were identified. After that, training attributes like texture, edge, and wavelet are extorted assistance dissimilar extortion approaches. will then be inputted PCA approach. similar replicated query images. At last, perform similarity measurement amid trained images attain distinguished image. addition, genuine general performance assesses projected technique assessed ICA also LDA techniques.