Developing a contactless palmprint authentication system by introducing a novel ROI extraction method
作者：
Highlights：
• A novel unrestricted, contactless palmprint image acquisition device is developed.
• A novel model-based ROI extraction method, robust to the small errors is proposed.
• A contactless palmprint DB is constituted by collecting 1752 images from 145 subjects.
• The images in DB have pose, rotation, position variations and cluttered backgrounds.
• The results achieved by the proposed system are encouraging: 0.277% EER and 99.488% RR.
摘要
•A novel unrestricted, contactless palmprint image acquisition device is developed.•A novel model-based ROI extraction method, robust to the small errors is proposed.•A contactless palmprint DB is constituted by collecting 1752 images from 145 subjects.•The images in DB have pose, rotation, position variations and cluttered backgrounds.•The results achieved by the proposed system are encouraging: 0.277% EER and 99.488% RR.
论文关键词：Contactless palmprint authentication system,Model-based ROI extraction,Palm direction estimation by nonlinear regression,Contactless palm database,AAM-based palm segmentation
论文评审过程：Received 25 August 2014, Revised 7 May 2015, Accepted 28 May 2015, Available online 24 June 2015, Version of Record 4 July 2015.
论文官网地址：https://doi.org/10.1016/j.imavis.2015.05.002
