手背静脉图像预处理算法研究  

Research on Pretreatment Algorithm of Hand Vein Image

作　　者：杨晓鹏[1] 蔡超峰[2] 潘珩[3] 赵玉霞[1] 

机构地区：[1]郑州大学第一附属医院医学装备部,河南郑州450052 [2]郑州轻工业学院电气信息工程学院,河南郑州45000 [3]郑州大学第一附属医院信息处,河南郑州450052

出　　处：《中国医疗设备》2013年第10期20-23,共4页China Medical Devices

基　　金：国家自然科学基金项目(61271146);河南省重点科技攻关项目(132102310380)

摘　　要：作为一种新颖的身份认证方式,手背静脉识别技术具有非常重要的现实意义和市场潜力。手背静脉识别通常基于静脉纹路的细节特征点对个人身份进行识别。为了准确地提取手背静脉图像的细节特征点,本文首先提取手背静脉图像中的有效区域并对其进行归一化;再利用子块部分重叠局部直方图均衡化算法对图像进行对比度增强并去除其中的斑点噪声和扫描噪声;最后分别采用NiBlack法和Hilditch算法对图像进行分割和细化处理,得到手背静脉图像的纹路信息。As an innovative authentication way, the hand vein recognition technology is provided with great realistic signiifcance and market potential. Minutiae of hand vein skeleton are usually employed in personal identiifcation recognition. To extract the minutiae of hand vein skeleton accurately, the effective area of hand vein image was extracted ifrstly and then normalized. Next, the partially overlapped sub-block histogram equalization （POSHE） was employed to enhance the contrast of the image followed by noise reduction. At last, the NiBlack algorithm was employed to segment the image and the Hilditch algorithm was employed to ifne treatment to obtain the hand vein skeleton.

关 键 词：身份认证 手背静脉识别 图像对比度 CCD摄像机 图像噪声 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]