Part based bit error analysis of iris codes

作者：

Highlights：

• Middle rings of the iris are more consistent than rings close to the pupil or limbus.

• Bit errors are uniform over different sectors if segmentation errors are eliminated.

• Iris code bit errors show the same trends for 3 databases and 3 recognition systems.

• Pupil size changes, resolution and angular resampling affect the bit errors most.

摘要

Highlights•Middle rings of the iris are more consistent than rings close to the pupil or limbus.•Bit errors are uniform over different sectors if segmentation errors are eliminated.•Iris code bit errors show the same trends for 3 databases and 3 recognition systems.•Pupil size changes, resolution and angular resampling affect the bit errors most.

论文关键词：Iris recognition,Iris code,Consistent bits,Error distribution

论文评审过程：Received 13 November 2015, Revised 24 March 2016, Accepted 16 May 2016, Available online 24 May 2016, Version of Record 9 June 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.05.022