FINGERPRINT IDENTIFICATION AND VERIFICATION USING MINUTIAE EXTRACTION FOR CRIME INVESTIGATION

Abstract

Crime has a negative impact on the socio- economic development of the world. Due to this Ethiopian federal police and other law enforcement agencies have the objective of effectively controlling crimes. These law enforcement agencies require assistance of scientific evidences during crime investigation. Fingerprint, as one of such scientific evidence, has an important scientific aid in the investigation of crime and administration of justice. Ensuring reliable minutiae extraction is one of the most important issues in automatic fingerprint identification and verification. The fingerprint identification and verification method is divided into four stages. The first is acquisition stage which captures the fingerprint image. The second is pre-processing stage which attempt for enhancement and binarization, of fingerprint images. In this work a novel method for fingerprint identification and verification is considered using a Fast Fourier Transform (FFT) to enhance the fingerprint image. The third stage is feature extraction, in this study the minutiae extractor methods are used to extract ridge ending and ridge bifurcation from thinned fingerprint image. The fourth stage is matching for fingerprint identification and verification. This is done by matching two minutiae points using minutiae matcher method in which similarity and distance measure are applied. We have used 300 fingerprint images for each of the 30 persons (ten fingerprints each) that are with criminals and innocent. From those images 85% of the dataset is used for training and 15% of the data set is used for testing. The experimental result demonstrates that the proposed technique is effective for the identification and verification of persons. The new developed method can successfully identify and verify the examined fingerprint images with an accuracy of 90.1%.