A Hybrid Algorithm for Fingerprint Segmentation from Dirty Glass and Metal Surfaces in Digital Image
fingerprint algorithm noise reduction original article:
In the world of digital image processing, feature extraction is one of the most significant and intrinsic elements, as data and information mining for deep learning or data analysis. Each object and concept feature extraction has its own issue and way, especially when our concept has got sensitive and delicate details, such as a fingerprint. Fingerprint detection and segmentation in a digital image are essential for criminal documentation or security as fingerprint matching or security archives in the document organisation. Finger patterns are usually visible and observable on transparent surfaces such as glass and metal, which have got a lot of noise and saturated light and spots. These impurities interfere with the analysis and recognition of fingerprints. In such a case, as fingerprint segmentation, we shall reduce the noise as much as possible and as long as the information is not damaged. We would have got a pure extracted fingerprint when we could separate finger patterns from the background. In this way, we need a particular combination of algorithms to extract finger patterns without damaging and disturbing the main information.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent