基于多重散射模型的手指静脉图像复原  

Restoration of finger vein image based on multiple scattering model

作　　者：杨金锋[1] 张锡瑞 YANG Jinfeng;ZHANG Xirui(Intelligent Signal and Image Processing Key Lab of Tianjin,CAUC,Tianjin 300300,China)

机构地区：[1]中国民航大学天津市智能信号与图像处理重点实验室,天津300300

出　　处：《中国民航大学学报》2020年第3期55-60,共6页Journal of Civil Aviation University of China

基　　金：国家自然科学基金项目(61073143)。

摘　　要：为提高手指静脉图像复原效果,更好地描述光子在皮肤组织中的传播特性,提出一种多重散射模型,并利用广义高斯分布函数(GGD)作为卷积核,近似刻画光子在皮肤组织传输衰减过程中的扩散现象。采用最大值滤波和各向异性扩散滤波估计背景光照,并结合拉普拉斯锐化和高提升滤波方法求取传输率图像,最终利用平均结构相似度指数进行质量评价。实验结果表明:该方法可有效提高手指静脉图像的复原质量。To improve the restoration quality of finger vein images and better describe the propagation characteristics of photons in skin tissue, a multiple scattering model is proposed by adopting GGD as convolution kernel to approximately describe the diffusion characteristics of photons during transmission and attenuation process in skin tissue. Then maximum filter and anisotropic diffusion filter are used to estimate the background lighting, and the transmission map is obtained by Laplacian sharpening and high lifting filter. MSSIM index is introduced to evaluate the quality of finger vein image restoration. Experiment results indicate that the proposed algorithm can effectively improve the restoration quality of finger vein images.

关 键 词：手指静脉图像 图像复原 多重散射 广义高斯分布函数 拉普拉斯锐化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]