IRIS IDENTIFICATION SYSTEM BASED ON RLM TEXTURE FEATURES (Published)
Among biometric identification technologies, iris recognition has attracted lots of attention because of uniqueness and long term stability. In this paper, a new iris recognition system based on texture has been proposed and used for identification of person. Texture features such as Run Length Matrix features (RLM) will be used for feature extraction. Inner and outer boundary will detect. Then iris region will divide into blocks and the importance degree of each block will assign as weights according to sigmoid function. Sixteen RLM features will be computed (eight features in each 0 and 90 directions). To evaluate the performance of proposed method, it applied to identify iris image from MMU ver.1 data set. Experimental results on 88 classes show that the new method give good recognition rates (99.3%) with smallest features vector as compared with other methods
Keywords: Biometric, Iris Recognition, Local Enhancement, RLM, Texture Feature