Palmprint Recognition Using Radon Transform and Relaxed Collaborative Representation Based Classifier

A relatively new technology, palmprint recognition, has evolved drastically over the recent years. It is currently considered to be one of the potential research topics due to the escalating need of biometric authentication based security systems. This paper presents a method of authentication of hand images by relaxed collaborative representation (RCR) model of classification. It makes use of the similarity as well as uniqueness of the associated features. By computing the radon transform of the images, the feature vectors of the images are computed. The coding of each feature vector is done over the dictionary that is linked with RCR and in order to compute the resemblance among features the inconsistency of coding vectors is reduced. The main aim is to develop an efficient system whose results can be implemented on any database.