基于Biohashing的指纹模板保护算法

Improved Biohashing Fingerprint Template Protection Algorithms

针对Biohashing指纹模板保护算法存在用户令牌泄露时识别性能严重退化的问题,提出了两种改进的Biohashing指纹模板保护算法.该算法在指纹数据预处理的基础上,采用可变的步长参数和滑功窗口产生固定大小的二值特征矩阵,减少了指纹数据特征值之间的关联性,离散化的非线性处理过程能够获得更大的密钥空间,有效提高了算法的安全性.理论分析和实验结果表明,改进算法具有更好的安全和识别性能.

Aimed at the problem that the Biohashing fingerprint template protection algorithm seriously degenerates when user＇s token is leaked, two improved Biohashing fingerprint template protection algorithms are proposed. On the basis of the preprocessing of fingerprint data, the variable step size parameter and sliding window are used to generate a fixed binary feature matrix, which can reduce the correlation between the eigenvalues of fingerprint data. The discretized nonlinear process can obtain a larger key space, which can effectively improve the security of the algorithm. Theoretical analysis and experimental results show that the improved algorithms have better security and identification performance.