研究目的
Investigating the phenomenon of template aging in iris recognition, specifically the increase in recognition error rate with increased time since enrollment.
研究成果
The study provides clear evidence of iris template aging, showing an increase in the false non-match rate with increased time since enrollment. This effect can be mitigated by enrolling multiple templates representing different degrees of pupil dilation and implementing rolling re-enrollment strategies.
研究不足
The study's findings are contingent on the specific sensor (LG 4000) and matcher (VeriEye SDK) used. The impact of other factors contributing to template aging, beyond pupil dilation, remains to be fully explored.
1：Experimental Design and Method Selection:
The study involved analyzing iris images acquired over a three-year period to assess the impact of time lapse on recognition accuracy. The VeriEye matcher was selected for its superior performance in preliminary tests.
2：Sample Selection and Data Sources:
Iris images were acquired from 2008 through 2011 using an LG 4000 sensor, involving 322 subjects with varying demographics.
3：List of Experimental Equipment and Materials:
LG 4000 iris sensor, VeriEye SDK (version 2.4) for iris matching.
4：4) for iris matching.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Short-time-lapse matches (within a few months) and long-time-lapse matches (between different years) were compared to assess the impact of time on recognition accuracy.
5：Data Analysis Methods:
The false non-match rate (FNMR) was analyzed as a function of the decision threshold, with bootstrap methods used to estimate confidence intervals for the change in FNMR.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
