首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于开关型非局部均值滤波的指静脉图像去噪
引用本文:杨 航,沈 雷,李 凡,吕葛梁.基于开关型非局部均值滤波的指静脉图像去噪[J].教育技术导刊,2020,19(1):262-266.
作者姓名:杨 航  沈 雷  李 凡  吕葛梁
作者单位:杭州电子科技大学 通信工程学院,浙江 杭州 310018
基金项目:杭州电子科技大学研究生科研创新基金项目(CXJJ2018029)
摘    要:为了解决带有皮裂纹特征的指静脉图像中干扰区域影响静脉特征提取的难题,提出一种基于开关型非局部均值滤波的指静脉图像去噪算法。该算法首先生成静脉图像像素点的Hessian矩阵,然后通过Frangi滤波对起皮干裂区域进行精准检测,最后对受干扰区域进行非局部均值滤波去噪。理论分析与实验结果表明,该算法与现有经典静脉去噪算法相比,充分利用了图像上其它区域的冗余信息,在实现去噪的同时,更好地保留了非受干扰图像区域的静脉细节部分,一定程度上克服了经典去噪算法对整个图像的过平滑问题。在认假率为0时,拒真率相比原结果降低了5.63%。

关 键 词:Hessian矩阵  Frangi滤波  非局部均值滤波  图像去噪  指静脉识别  
收稿时间:2019-04-08

Finger Vein Image Denoising Based on Switching Non-local Mean Filtering
YANG Hang,SHEN Lei,LI Fan,LV Ge-liang.Finger Vein Image Denoising Based on Switching Non-local Mean Filtering[J].Introduction of Educational Technology,2020,19(1):262-266.
Authors:YANG Hang  SHEN Lei  LI Fan  LV Ge-liang
Institution:School of Communication Engineering,Hangzhou Dianzi University,Hangzhou 310018,China
Abstract:In order to solve the problem that the disturbed area in the finger vein image with the peeling cracking feature affects the vein feature extraction, a finger vein image denoising algorithm based on switching non-local mean filtering is proposed. The algorithm first generates the Hessian matrix of the pixel points of image. And then the peeling and cracking area is accurately detected by Frangi filtering. Finally,and the disturbed region is denoised by non-local mean filtering. Experimental results and theoretical analysis indicate that compared with the existing classical vein denoising algorithm, the proposed algorithm makes full use of the redundant information of other areas on the image, and better preserves the vein detail of the non-disturbed image area. To a certain extent, it overcomes the problem that the classical denoising algorithm is too smooth for the entire image. In the case of the false acceptance rate of 0,the false rejection rate is reduced by 5.63% compared to the original result.
Keywords:Hessian matrix  Frangi filtering  NL-means filtering  image denoising  finger vein recognition  
点击此处可从《教育技术导刊》浏览原始摘要信息
点击此处可从《教育技术导刊》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号