基于shape context的指纹图像识别研究  

Fingerprint image recognition research based on shape context

作　　者：胡晓霞[1] 郑三婷 HU Xiao-xia;ZHENG San-ting(North Institute of Information Engineering Xi’an Technological University,Xi’an 710025,China;Xi’an International University,Xi’an 710077,China)

机构地区：[1]西安工业大学北方信息工程学院,陕西西安710025 [2]西安外事学院,陕西西安710077

出　　处：《电子设计工程》2019年第17期122-126,共5页Electronic Design Engineering

基　　金：国家自然科学基金(61671362)

摘　　要：针对智能身份管理系统中的指纹图像识别问题,提出了一种在形状上下文特征提取算法的前提下,基于多种现有度量算法相结合的联合度量策略的图像识别机制。首先,描述了基于形状上下文的指纹图像的特征提取过程;其次,在分析现行图像识别方法的基础上,提出了基于经典的欧式距离度量算法和TPS薄板样条法进行的联合度量策略的总体思路;而后讨论了基于联合度量策略机制的识别算法在图像识别中的可行性及识别精度,其识别率达到87.5%,并具备较高的稳定性;最后,将该算法应用在实际的指纹图像的识别系统中,指纹图像识别率达到82.5%,实验结果表明所提出的方法是有效的。A new image recognition mechanism based on a combination of several existing measurement algorithms is proposed under the premise of shape context feature extraction algorithm to solve the problem of fingerprint image recognition in intelligent identity management system. Firstly,the feature extraction process of fingerprint image based on shape context is described. Secondly,based on the analysis of the current image recognition methods,the general idea of the combined measurement strategy based on the classical Euclidean distance measurement algorithm and TPS thin plate spline method is proposed. Then,the feasibility and accuracy of the recognition algorithm based on the joint measurement strategy mechanism are discussed,the recognition rate of fingerprint image is 87.5%.And this algorithm has high stability. Finally,the algorithm is applied to the actual recognition system of fingerprint image. The recognition rate of fingerprint image is 82.5%. Experimental results show that our proposed method is effective.

关 键 词：指纹识别 形状上下文 薄板样条曲线 联合度量策略 

分 类 号：TN911[电子电信—通信与信息系统]