应用于嵌入式设备的指纹图像质量估计方法  

Fingerprint Image Quality Estimation Method Applied to Embedded Devices

作　　者：刘小强 袁国顺[1] 乔树山[1] LIU Xiaoqiang;YUAN Guoshun;QIAO Shushan(Institute of Microelectronics,Chinese Academy of Sciences,Beijing 100029,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院微电子研究所,北京100029 [2]中国科学院大学,北京100049

出　　处：《湖南大学学报（自然科学版）》2019年第12期9-15,共7页Journal of Hunan University:Natural Sciences

基　　金：国家自然科学基金资助项目(61474135);国家高技术研究发展计划资助项目(2015AA042605)~~

摘　　要：现有的指纹图像质量估计算法时间复杂度较高,嵌入式设备的计算资源有限,导致时间复杂度较高的指纹图像质量估计算法在这些设备上应用较为困难,为了解决这一问题,提出了应用于嵌入式设备的指纹图像质量估计方法.该方法首先对梯度场、方向场和频率场进行估计,通过对高质量指纹图和纹理图像的分析,得到了指纹图像质量索引与梯度、方向和频率之间的关系.利用这一关系衡量梯度、方向和频率的准确性,进而作为指纹图像质量索引来表征指纹图像质量.实验结果表明,该方法能够准确地产生质量索引,可以很好地对指纹质量优劣区域进行区分,在保证指纹认证系统性能的前提下,显著降低了指纹质量估计算法时间复杂度.The existing fingerprint image quality estimation algorithm has high computational complexity.The computing resources of embedded devices are limited,and fingerprint image quality estimation algorithm with high computational complexity is difficult to apply on these devices.In order to solve this problem,a method of fingerprint image quality estimation method applied to embedded devices is proposed.First,the method estimates the gradient field,the direction field and the frequency field,and the relationship among the fingerprint image quality,gradient,direction and frequency is obtained by analyzing the high-quality fingerprint image and texture image.This relationship is used to measure the accuracy of the estimated gradient,direction and frequency,and then it is made as the fingerprint image quality index to characterize the quality of fingerprint image.The experimental results show that the method can accurately generate the fingerprint image quality index.The method can distinguish the good and bad regions of the fingerprint image very well.Under the premise of ensuring the performance of the fingerprint authentication system,the computational complexity of the fingerprint quality estimation algorithm is significantly reduced.

关 键 词：指纹认证 图像质量 方向估计 频率估计 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]