Search

Recently, biometric was being integrated with cryptography (crypto-biometric system) to alleviate the limitations of the biometric or cryptography system. However, the main shortcoming of cryptography is poorly-chosen or forgotten password while challenges with biometrics include interclass similarities in the feature sets used to represent traits. In this work, a combination of cryptography and bimodal biometric was developed, an Advanced Encryption Standard based Fast Fourier Transform (AES-FFT) was developed and used as the cryptography technique. Hence, an attempt was made to develop an improved access control system using an enhanced bimodal bio-cryptography. Biometric features was extracted from individual face and iris after application of suitable preprocessing techniques for each modality using Principal Component Analysis (PCA) while cryptography key was generated using fused features from the face and iris by Advanced Encryption System based Fast Fourier Transform (AES-FFT). The two captured biometric data at acquisition module via webcam were subjected to appropriate pre-processing and feature extraction module. The features extracted were fused at feature level using weighted average and optimal features were selected using genetic programming (GP). The classification technique used was Support Vector Machine (SVM).