Single Image Ear Recognition Using Wavelet-Based Multi-Band PCA

Abstract

Principal Component Analysis (PCA) has been successfully used for many applications, including ear recognition. This paper presents a 2D Wavelet based Multi-Band PCA (2D-WMBPCA) method, inspired by PCA based techniques for multispectral and hyperspectral images, which have shown a significantly higher performance to that of standard PCA. The proposed method performs 2D non-decimated wavelet transform on the input image dividing the image into its subbands. It then splits each resulting subband into a number of bands evenly based on the coefficient values. Standard PCA is then applied on each resulting set of bands to extract the subbands eigenvectors, which are used as features for matching. Experimental results on images of two benchmark ear image datasets show that the proposed 2D-WMBPCA significantly outperforms both the standard PCA method and the eigenfaces method.

Human ear print recognition based on fusion of difference theoretic texture and gradient direction pattern features

“…Zarachoff et al [16] presented a two-dimensional wavelet relying on multi-band PCA (2DWMBPCA) technique, inspired through PCA based on technical for hyperspectral and multispectral images, which showed significantly higher performance than standard PCA. The suggested technique performs a non-destructive twodimensional wavelet transformation on the input image to split the image into its sub-bands.…”

Non-Decimated Wavelet Based Multi-Band Ear Recognition Using Principal Component Analysis

“…To the authors' knowledge, no similar techniques that utilize wavelets have been reported in the literature. This has inspired the authors to propose a single image, PCA based method for ear recognition, called 2D Wavelet based Multi-Band PCA (2D-WMBPCA), which was originally introduced in EUSIPCO in [17]. Unlike the aforementioned PCA based methods, the proposed technique does not require the images to be projected into a common eigenspace.…”