指纹纹线距离估计:算法与性能评价  

FINGERPRINT RIDGE DISTANCE ESTIMATION: ALGORITHM AND PERFORMANCE EVALUATION

作　　者：尹义龙[1] 孟祥旭[1] 

机构地区：[1]山东大学计算机科学与技术学院,济南250100

出　　处：《模式识别与人工智能》2004年第2期190-195,共6页Pattern Recognition and Artificial Intelligence

基　　金：山东大学校青年科学基金(No.2003-27);山东大学计算机科学与技术学院学科建设(No.10000052182441)资助项目;安徽省教育厅自然科学研究(No.2002KJ234)资助项目

摘　　要：指纹纹线距离的可靠估计对自动指纹识别系统的实现具有重要的意义.本文利用传统的谱分析方法在频域实现了对指纹图像纹线距离的估计,提出了一种基于统计窗的纹线距离估计新的方法,在空间域实现了纹线距离估计.针对纹线距离估计方法缺少性能评价标准的现状,定义直接估计出纹线距离的区域占整个指纹有效面积的比例(Direct Estimation Ratio,DER)和估计精度(Estimation Accuracy,EA),提出使用DER、EA和时间消耗(TimeConsuming,TC)作为评价纹线距离估计方法的性能指标.实验结果表明,在模式区附近等纹线方向变化剧烈的区域,谱分析方法可以更为可靠地实现对纹线距离的估计,而在纹线方向变化平缓的区域,统计窗方法对图像质量表现出更好的鲁棒性,可以取得更好的效果.谱分析方法的DER、EA、TC分别为46.7%、84%、0.42秒,而统计窗方法的DER、EA、TC分别为66.4%、95O、0.29秒,统计窗方法在整体性能上要优于谱分析法.It is important to estimate ridge distance reliable for automatic fingerprint identification system. Traditional spectral analysis method for ridge distance estimation is realized in frequency domain and a new method, which is called statistical window method, is brought out to estimate ridge distance in spatial domain. Aiming at the problem of lack of standards to evaluate the performance of ridge distance estimation methods, Direct Estimation Ratio (DER) and Estimation Accuracy (EA ) are defined and are used along with Time Consuming (TC ) to evaluate the performance of the two methods. Experimental results indicate that DER, EA, TC with spectral method are 46. 7%, 84%, 0. 42s, while DER, EA, TC with statistical method are 66. 4%, 95%, 0. 29s. The average performance of statistical window method is superior to that of spectra1 analysis method.

关 键 词：指纹 纹线距离 谱分析 统计窗 基线 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]