Enhance Gabor filter based fingerprint classification based on Termination and Bifurcation extraction

Abstract

In forensic science, distortions in fingerprint images are a major problem because they make matching methods less accurate and of lower quality. To achieve reliable identification, fingerprint scans must be of high quality. In this project, a Gabo filter was used to improve image quality and make it easier to see small details. Additionally, a study was conducted to compare the success of two new fingerprint classification methods. The first clustering method used termination features. Fuzzy-C-Min clustering operators were used, and we adopted two experiments. The first experiment adopts the Terminations features are adopted for attending the clustering process. The second experiment adopts the Bifurcations features are adopted for attending the clustering. The results showed that the second experiment was better at obtaining a higher level of classification accuracy. In other words, making automated fingerprint analysis tools at crime scenes work better.