Multi-band PCA based ear recognition technique

Abstract Principal Component Analysis (PCA) has been successfully applied to many applications, including ear recognition. This paper presents a Two Dimensional Multi-Band PCA (2D-MBPCA) method, inspired by based techniques for multispectral and hyperspectral images, which have demonstrated significantly higher performance that of standard PCA. The proposed method divides the input image into number images on intensity pixels. Three different methods are used calculate pixel boundaries, called: equal size, histogram, greedy hill climbing techniques. Conventional is then resulting extract their eigenvectors, as features. optimal bands was determined using intersection features total eigenvector energy. Experimental results two benchmark datasets demonstrate 2D-MBPCA technique outperforms single up 56.41% eigenfaces 29.62% with respect matching accuracy from datasets. Furthermore, it gives very competitive those learning at fraction computational cost without need training.