Cloud-based human authentication through scalable multibiometric image sensor fusion

The privacy and safety of confidential information throughout various programs are greatly enhanced by cloud-based human verification. Traditional authentication techniques frequently rely on a single biometric property, which can be vulnerable to assaults like spoofing. This work offers an extensible multibiometric image sensor fusion method for cloud-based human authentication using an improved particle swarm optimization-based Advanced Encryption Standard (IPSO-AES) cryptography framework. The IPSO-AES model that has been suggested combines several biometric features, including fingerprint, palm print, and facial characteristics, which have been recorded by a flexible multibiometric image detector. To improve the quality and dependability of the collected biometric information, preprocessing is applied. The AES method, which is applied to encrypt the biometric information before transfer to the cloud system, is then optimized using the IPSO technique. Using a wide range of trials, the suggested model is assessed in terms of several metrics and contrasted with current authentication techniques. The experimental findings show that the suggested encryption model performs better than other alternatives with regard to authentication precision, security, and scalability. The concept is promising for actual cloud-based human verification scenarios where anonymity and safety are crucial.