Robust Image Recognition Based on a New Supervised Kernel Subspace Learning Method

Abstract

Image recognition is a term for computer technologies that can recognize certain people, objects or other targeted subjects through the use of algorithms and machine learning concepts. Face recognition is one of the most popular techniques to achieve the goal of figuring out the identity of a person. This study has been conducted to develop a new non-linear subspace learning method named “supervised kernel locality-based discriminant neighborhood embedding,” which performs data classification by learning an optimum embedded subspace from a principal high dimensional space. In this approach, not only is a nonlinear and complex variation of face images effectively represented using nonlinear kernel mapping, but local structure information of data from the same class and discriminant information from distinct classes are also simultaneously preserved to further improve final classification performance. Moreover, to evaluate the robustness of the proposed method, it was compared with several well-known pattern recognition methods through comprehensive experiments with six publicly accessible datasets. In this research, we particularly focus on face recognition however, two other types of databases rather than face databases are also applied to well investigate the implementation of our algorithm. Experimental results reveal that our method consistently outperforms its competitors across a wide range of dimensionality on all the datasets. SKLDNE method has reached 100 percent of recognition rate for Tn=17 on the Sheffield, 9 on the Yale, 8 on the ORL, 7 on the Finger vein and 11on the Finger Knuckle respectively, while the results are much lower for other methods. This demonstrates the robustness and effectiveness of the proposed method.