Fusion in multimodal biometric using iris and ear

Unimodal biometric systems have to contend with a variety of problems such as noisy data, intra-class variations, spoof attacks, and unacceptable error rates. Some of these limitations can be addressed by multimodal biometric systems that integrate multiple sources of human information to identify as well to provide ultra secure system for the information. The main objective of the proposed system is to analyze the performance of two traits, namely, ear and iris, individually and combined them by applying score level fusion technique. Ear and Iris Recognition system was built by extracting their features using Principal Component Analysis (PCA) technique by determining the Eigen vectors for dimensionality reduction without information loss. The similarity between the test data and the training set is measured and combined together using sum rule based score level fusion method. This proposed system is implemented to study and analyze the performance of multi traits during fusion. The fusion work results to 95% success rate, which is higher rather than a Unimodal system.