基于多角度旋转积分图的手背静脉身份识别

Dorsal Hand Vein Recognition Based on Multi-angle integration

摘要:

随着生物特征识别技术水平的飞速发展,手背静脉识别也广泛运用于各个领域。由于采集终端硬件设备和采集环境的差异,会降低识别的准确性。针对手背静脉图像在亮度,旋转,尺寸等方面造成的影响,,提出了基于多角度旋转积分图的和离散余弦变换的手背静脉图像识别方法,并进行参数优化。本文结合梯度增强的静脉图像分割方法,选取最佳角度间隔做旋转积分运算,然后通过二维离散余弦变换(DCT)截取最佳特征矩阵用做分类识别,识别率超过99.9％。 实验通过对比其它传统算法对手背静脉图像的识别效果来验证本文特征提取方法的可行性和优越性。

Abstract:

With the continuous rapid development of biometric technology, dorsal hand vein recognition technology has been more and more widely used in all fields. Because of the influence of the different hardware conditions and environment, the recognition rate is reduced. For the problem of brightness, rotation, scales of the dorsal hand vein images, we propose a method of dorsal hand vein recognition based on multi-angle rotation integration and discrete cosine transform, and the parameter is optimized. We combine a method of image-binary-segmentation which is based on the gradient enhancement, select the best angle interval for the rotation integration, intercept best feature matrix for classification recognition after two-dimensional discrete cosine transform, the recognition rate is more than 99.9%. The experiment verifies the feasibility and superiority of the method proposed in this paper by comparing the recognition effects of other traditional algorithm of dorsal hand vein images.