E-Authentication System Using Biometrics

Concerns about thumbprint biometric security have grown in recent years This problem is brought on by technological developments in evading and exploiting. This has led to a demand for a supplementary sheltered raised area for recognition. In this post, we developed an interactive neural network based on deep learning as a pre-authentication filtering mechanism to eliminate unwanted or harmful fingerprints. The suggested approach significantly improves security and accuracy since deep learning enables systems to become more successful at identifying and eliminating false positives by regularly using test data and teaching themselves. This study describes the development of an original, secure fingerprint verification system that can use a handprint image as an input. Pre-trained Inception model by Google for profound erudition applications is used to pre-verify the input before it is sent to a biometric based technique for user verification. The outcomes are then compared to current systems.