Fingerprint Image Enhancement Based on Mean Adjustment, Feature Extraction & Matching

Accurate involuntary personal identification is critical in a variety of applications in our electronically interconnected society. Biometrics, which mentions to identification based on physical or behavioural characteristics, is being progressively adopted to provide positive identification with a high degree of confidence. Among all biometric techniques, fingerprint-based authentication schemes have establishedmost attention because of long history of fingerprints and their extensive use in forensics. Fingerprints are a great source for identification of individuals. Fingerprint recognition is one of the forms of biometric identification. However obtaining a decent fingerprint image is not always easy. So, fingerprint image must be pre-processed before matching. The main objective of this work is to design an image matching algorithm which is applied to any image for matching. For efficient enhancement and feature extraction procedures, the segmented structures must be void of any noise. A pre-processing method containing of field orientation, ridge frequency approximation, filtering, separation and enhancement is performed. The attained image is applied to a thinning algorithm and subsequent minutiae removal. The organization of image pre-processing and minutiae extraction is deliberated. The simulations are performed in the MATLAB environment to evaluate the performance of the implemented algorithms. MATLAB provides a useful environment for these developments. Results and observations of the fingerprint images are presented at the end.