一种指纹图像简化预处理与特征提取算法实现  

ASimplified Preprocessing and Feature Extraction Algorithm for Fingerprint Image

作　　者：罗海宁 王建英[2] LUO Haining;WANG Jianying(School of Traffic and Transportation,Lanzhou Jiaotong University,Lanzhou 730070,China;School of Electronic Information and Automation,Tianjin University of Science and Technology,Tianjin 300457,China)

机构地区：[1]兰州交通大学交通运输学院,兰州730070 [2]天津科技大学电子信息与自动化学院,天津300457

出　　处：《自动化与仪器仪表》2022年第1期26-29,34,共5页Automation & Instrumentation

摘　　要：自动指纹识别技术已经步入一个新的时期,现有的自动指纹识别技术在愈加完善和精确的同时也愈加复杂。在保证精确的前提下,有效提高计算效率,以更好地适应移动互联网时代背景下对自动指纹识别技术的要求,成为了一个亟待解决的问题。指纹图像预处理和特征提取中的算法优化对于指纹识别技术的精度和速度具有关键的影响。为了改善指纹图像预处理过程中的计算效率问题,在使用规格化算法进行对比度调节的基础上,针对指纹采集时产生的噪声使用8方向的方向滤波算法,并在之后使用局部自适应的阈值算法进行二值化处理。在图像细化环节,使用形态学快速细化算法替换传统的OPTA细化算法进行细化。在对细节特征进行提取后,对图像边缘的伪特征切割、去除并使用欧式距离作为判断依据以去除伪特征点。结果表明,此种算法能够有效解决计算效率的问题。Automatic fingerprint identification technology has entered a new era,the existing automatic fingerprint identification technology is more perfect and accurate,but also more complex.Under the premise of ensuring accuracy,it is an urgent problem to improve the computational efficiency to better adapt to the requirements of automatic fingerprint identification technology in the mobile Internet era.The algorithm optimization of fingerprint image preprocessing and feature extraction has a key impact on the accuracy and speed of fingerprint identification technology.In order to improve the computational efficiency of fingerprint image preprocessing,on the basis of using the normalized algorithm to adjust the contrast,the 8-direction direction filtering algorithm is used for the noise generated during fingerprint acquisition,and then the local adaptive threshold algorithm is used for binarization.In the process of image thinning,morphological fast thinning algorithm is used to replace the traditional OPTA thinning algorithm.After the detail features are extracted,the pseudo features of the image edge are cut and removed,and the Euclidean distance is used as the judgment basis to remove the pseudo feature points.The results show that this algorithm can effectively solve the problem of computational efficiency.

关 键 词：指纹识别 预处理 方向滤波 形态学算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]