摘要
针对手形的特点和现有手形认证方法的不足,提出了一种基于曲线拟合的手形生物特征认证新算法·该算法使用手指轮廓拟合曲线的系数作为手形的特征,使用曲线距离函数进行匹配认证,进一步导出基于曲线系数进行求解的简化方法·实验表明,该算法的认证错误接收率和错误拒绝率之和达到1%以下;与现有的手形认证方法相比,该算法在认证的准确率、鲁棒性和运算量方面具有良好的综合性能·
A new hand shape biometric verification method based on curve fitting is proposed, which uses the curve fitting coefficients of the finger contour as features. Curve distance is defined to compare different finger curves and make verification decision. In addition a simplified calculation method based on the curve fitting coefficients is presented. It＇ s shown by experimentation that the sum of false acceptance rate and false rejection rate is below 1 percent. Compared with the existing methods, this method has a high tradeoff performance on the accuracy, robustness and efficiency.
出处
《计算机研究与发展》
EI
CSCD
北大核心
2005年第11期1870-1875,共6页
Journal of Computer Research and Development
关键词
生物特征识别
手形生物认证
曲线拟合
曲线距离
biometrics
hand shape biometric verification
curve fitting
curve distance
