Kernel PCA enabled bit-string representation for minutiae-based cancellable fingerprint template
作者：
Highlights：
• A fixed-length binary representation of fingerprint using kernel PCA is proposed.
• Performance after applying KPCA drops slightly.
• Performance of the final bit-string is excellent, achieving average EER of 0.09%.
• The method is strong against inverse and linkage attacks but not masquerade attack.
• The computational complexity of the proposed scheme is high.
摘要
Highlights•A fixed-length binary representation of fingerprint using kernel PCA is proposed.•Performance after applying KPCA drops slightly.•Performance of the final bit-string is excellent, achieving average EER of 0.09%.•The method is strong against inverse and linkage attacks but not masquerade attack.•The computational complexity of the proposed scheme is high.
论文关键词：Cancellable fingerprint,Multi-line code (MLC),Kernel principal components analysis (KPCA),Fixed-length representation
论文评审过程：Received 21 April 2014, Revised 23 September 2015, Accepted 24 September 2015, Available online 9 October 2015, Version of Record 27 November 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.032
