基于小波分析的方向滤波算法研究  

Direction-Map Based Fingerprint Enhancement Algorithm in Wavelet Domain

作　　者：王建英[1] 李临生[1] 田启川[1] 

机构地区：[1]太原科技大学电子信息工程学院,太原030024

出　　处：《科学技术与工程》2007年第12期2872-2875,共4页Science Technology and Engineering

摘　　要：针对采集到的指纹库中存在指纹过湿,以及存在高频噪声的问题,提出了一种基于小波分析的方向滤波算法。该算法在小波域利用低频系数图估计纹线方向,抑制了高频噪声的影响,实现了小波域低频图的方向滤波。最后利用小波逆变换实现方向滤波后的指纹图像的重构,试验表明,该算法对低质量指纹图像的增强效果明显。There exist some wet and high-frequency noise fingerprint in large fingerprint databases. A low quality fingerprint enhanced algorithm was designed in wavelet domain. The algorithm uses low-frequency coefficients estimated the direction of the ridge, so the high-frequency noise impact was reduced. Each direction coefficient images were enhanced based on direction-map and further reconstruction. The simulation results on some typical images show that algorithm has better enhanced performance and lower computation burden .

关 键 词：纹线方向 方向图 方向滤波 小波分析 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]