This paper presents a comparison study of Gaussian Mixture Models for fingerprints image duplication and analysis. It also presents a new probabilistic Parametric Gaussian Mixture Model(GMM). The system is built around the likelihood ratio test for verification, using simple but
effective GMMs for likelihood functions and a form of Bayesian adaptation to derive the models. The Computer simulation show that the developed new algorithms have the most optimal performance as compared to state of art algorithms GMMs, Generalized GMMs, Finite Bayesian learning for GMMS,
Texture Synthesis and Improved Adaptive Algorithm. The performance of the presented algorithm was evaluated by Bovik Index, Entropy and Mean Square Error.