针对手指静脉的身份识别问题，并结合手指静脉具有丰富纹理信息的特点，提出了一种基于小波灰度曲面的近红外手指静脉识别方法。采用直方图均衡化方法对原始静脉图像的感兴趣区域进行灰度调整，再用小波分解降维，提取降维后具有不同分辨率的图像，构建待匹配图像。将两幅待匹配图像中的对应像素值相减, 得到灰度差曲面。求出该灰度差曲面的方差, 将其作为衡量两个手指静脉特征曲面之间距离的依据, 并据此判定两个静脉是否来自同一个手指。应用该方法在国内和国外两个图库中使用典型和流行方法进行了对比实验，结果表明，提出的方法用Haar 小波降维后可获得具有不同分辨率的图像，在两个图库上的最低等误率(EER)分别为0%和4.6281%，识别时间仅为0.061 s 和0.0502 s。该算法具有一定的优势和可行性，且准确性高、安全保密性好、运行速度快。

Aiming to identify the finger vein and considering the rich texture characteristics of the finger vein, a near infrared finger vein recognition approach based on wavelet grayscale surface matching is proposed. The region of interest of the original image is adjusted by using the histogram equalization, the different resolution images are extracted after decomposition, and the images for matching are constructed. The gray difference surface is obtained by computing the gray difference of two pixels from two different images. The variance is calculated by using the gray difference surface, and is considered as the distance between two feature surfaces of the finger vein images, and the result is used to determine whether the two finger vein images are from the same finger. The comparison experiments are performed with the typical and popular approaches on two databases. The experimental results show that the lowest equal error rate (EER) is 0% and 4.6281% respectively, and the recognition time is only 0.061 s and 0.0502 s, respectively, when the different resolution images are extracted after Haar wavelet decomposition. The superiority and feasibility of the proposed approach is indicated , and high accuracy, good security and fast running speed of the approach are exhibited.