主曲线在复杂指纹图像特征提取中的应用

摘要

在自动指纹识别系统中，特征抽取是关键步骤之一。主曲线具有自相合特性，对模式特征能够进行很好的描述，并能够有效维持结构信息。因此，选用推广的多边形主曲线算法并加以改进来提取指纹主曲线，并在此基础上进一步实现指纹特征提取和伪特征检测。实验结果表明，该算法能够在短时间内获得更好的指纹骨架，指纹特征提取的准确率也较高。%In automated fingerprint recognition system , feature extraction is one of the key procedures .Principal curve has the property of self-consistency , can well depict the pattern feature and can effectively keep the structure information .Therefore , we choose the promoted polygonal principal curve algorithm and improve it to extract the principal curve of the fingerprint , and further implement on this basis the fingerprint extraction and pseudo feature inspection .Experimental results demonstrate that the algorithm in this paper can get better fingerprint skeleton in short time period , the accuracy rate of fingerprint feature extraction is higher as well .