Published Paper Details:

The analysis reviews the combination of Bayesian classifiers with Gabor filters together with Support Vector Machines (SVMs) for enhancing secure face authentication throughout cloud-based systems. The paper stresses the importance of Gabor wavelets because these filters extract face features and retain their discrimination properties despite lighting and positioning changes. The SVM method provides reliable classification functions and it enables separation of face entities from non-face elements in simplified representation areas. A study of Bayesian approaches enables the enhancement of authentication system reliability by dealing with probabilistic uncertainties present in encrypted facial templates. The framework combines those techniques with principles of homomorphic encryption to establish protected storage and processing of biometric data in cloud systems that guarantee privacy. This paper examines computational complexity alongside adversarial vulnerabilities through an analytical assessment before providing security methods and efficiency enhancement solutions. The method combines these technologies to build a solution for secure medical data manipulation that can expand within decentralized cloud networks.