摘要： 针对光源、手背厚度及水纹对手背静脉图像的噪声影响问题，提出一种新的手背静脉特征提取方法，包括最近邻图像插值缩放、Canny边缘检测以及一种改进的基于灰度直方图加权并顾及像素空间信息的FCM算法。实验结果证明，应用改进的FCM算法能减少图像分割迭代次数，较好地滤除噪声干扰。

关键词: Canny算子, FCM算法, 图像分割

Abstract: In order to overcome the effects which light source, the thickness of the hand skin or water refraction take on the hand vein picture, this paper proposes a novel algorithm to extract the hand vein features. It includes the nearest-neighbor interpolation, Canny operator and an improved FCM algorithm considering the information of pixels and based on picture grayscales. Experimental results indicate the new method can reduce the iteration times of image segmentation, and can filter the noise effectively.

Key words: Canny operator, FCM algorithm, image segmentation

中图分类号: 

• TP391