Enhanced Fingerprint Recognition by Reference Auto-correction with FCM-CBIR strategy

Abstract

Nowadays fingerprint recognition becomes an important biometric trait for authenticating and identifying individuals. Various researchers found that there is a change in the fingerprint images due to the variation in temperature. Hence in this paper, an effective fingerprint recognition system is developed to recognize the fingerprint images varied due to environmental changes like temperature. Hence, we propose FCM-CBIR technique for the identification of fingerprint during change of temperature. In this, clustering is performed using fuzzy $\mathbf{c}$ means clustering algorithm and the image retrieval process is performed using CBIR Content Based Image Retrieval. By using this proposed method, the unrecognized fingerprint due to temperature changes has been identified. The performances are measured using the various fingerprint images collected from real time environment.