Ear Identification Based on KICA and SVM

This paper gives a research in ear identification. After giving an introduction about Independent Components Analysis (ICA), the paper put forward an improved Kernel Independent Components Analysis (KICA) which can be described as a combination of KPCA and ICA to extract features, And use Support Vector Machine (SVM) with Gaussian radial basis function (GRBF) for ear classification. The experiments results show the method in the paper gives a high recognition rate compared to ICA method.