Three Way Access Control Using Biometrics and Data Mining Algorithm

Abstract

In recent years, security and privacy are very important to keep our data safely. The normal cryptographic key generation for encryption and decryption is not highly secured in this modern world. This work introduced biometric concept for cryptography key generation for authentication process. Behavioral and physical characteristics are measured to identify individuals by biometrics. As every individual has a unique identity like IRIS, finger prints, palm prints, retinal, face, etc. It is difficult to strengthen the security process using single modality biometric system as it dealt with decision making. So, it is important to introduce multimodal biometric system which uses dual finger print, face and retina images for creating a multimodal template. This work also adopted SVM classifier for performing verification as well as validation. The given user biometric feature vector is verified using the existing binary digit of the individual personal key for acceptance or rejection. The multimodal template can be further used to generate the cryptography key. This proposed approach using weights can be helpful to simplify the key generation process but also increases the security of the data as well as privacy of users.