基于梯度法的指纹方向场的FPGA实现  

FPGA implementation of fingerprint orientation field based on gradient method

作　　者：张宜尧 徐银森 李健儿 陈勇[1] 赵建明[1] ZHANG Yiyao;XU Yinsen;LI Jianer;CHEN Yong;ZHAO Jianming(University of Electronic Science and Technology of China,Chengdu 610054,China;Sichuan Suining lippxin Microelectronics Co.,Ltd.,Suining 629000,China;Sichuan Uptech Technology Co.,Ltd.,Suining 629201,China)

机构地区：[1]电子科技大学,四川成都610054 [2]四川遂宁市利普芯微电子有限公司,四川遂宁629000 [3]四川上特科技有限公司,四川遂宁629201

出　　处：《电子设计工程》2022年第20期23-27,33,共6页Electronic Design Engineering

摘　　要：文中提出了一种基于梯度法的指纹方向场的FPGA实现方式,该方案主要由Sobel算子模块及区域方向计算模块组成。通过Sobel算子模块得到指纹图像像素点的一阶梯度;采用区域方向计算模块对指纹图像进行分块,并利用方差法计算得到各指纹分块区域的方向值。通过对所述模块进行综合及仿真,验证其功能的实现。结合方向场质量评价方法,分析不同分块区域大小对所提取方向场质量的影响。通过与传统方向模板法得到的实验结果进行对比评估,由实验结果可知,该方案可为后续指纹图像处理提供强坚固性的纹线方向特征信息。This paper presents a FPGA implementation of fingerprint direction field based on gradient method,which is mainly composed of Sobel operator module and region direction calculation module.The first step degree of fingerprint image pixel is obtained by Sobel operator module;The fingerprint image is divided into blocks by using the region direction calculation module,and the direction value of each fingerprint block area is calculated by using the variance method;Through the synthesis and Simulation of the module,the realization of its function is verified. Combined with the quality evaluation method of directional field,the influence of different block size on the quality of extracted directional field is analyzed. By comparing with the experimental results obtained by the traditional orientation template method,it is concluded that this scheme provides strong and firm ridge direction feature information for subsequent fingerprint image processing.

关 键 词：FPGA 方向场 SOBEL算子 质量评价 

分 类 号：TN791[电子电信—电路与系统]