A study on phalangeal joint reference line detection for finger vein images

• DOI number:10.1109/FSKD.2018.8686986
• Affiliation of Author(s):西南交通大学
• Journal:2018 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)
• Place of Publication:Huangshan, China
• Key Words:image processing, finger vein image, region of interest, reference line detection, sliding window
• Abstract:Region of interest extraction is crucial for both the finger vein recognition system and the finger vein based cryptography system. Accurate phalangeal joint reference line detection is a key step in many ROI extraction methods for finger vein images and the position of the reference line is usually detected by a sliding line or rectangle window. Initiated by the brighter irregular shape of the finger joint region, an ellipse sliding window is used in our proposed method. In order to find out the most suitable ellipse window, a series of ellipse windows with different width is studied in our experiment. Based on our proposed contour similarity distance evaluation method, the performances of finger joint reference line detection by various ellipse windows are evaluated. And the experimental results show that the ellipse whose ratio of width to height is 0.85 has the best overall performance.
• Co-author:Dongming Tang
• First Author:Mingwen Wang
• Indexed by:Academic papers
• Correspondence Author:Mingwen Wang
• Document Code:20165203196233
• Discipline:Engineering
• First-Level Discipline:Computer Science and Technology
• Volume:76
• Issue:13
• ISSN No.:978-1-5386-8097-1
• Translation or Not:no
• Date of Publication:2018-07-28