中文摘要：

针对灰度不均匀、对比度低、边缘信息较弱的图像中的手指静脉纹路提取问题，提出了一种基于Hessian矩阵的手指静脉图像分割方法。该方法首先将高斯滤波器的二阶导数与原图像卷积得到了各像素点的Hessian矩阵，通过Hessian矩阵的迹初次滤除了非静脉区的像素点，接着求出了余下像素点Hessian矩阵的特征值，利用静脉区特征值所要满足的条件二次滤除非静脉区像素点，最后选取了多尺度下静脉区各像素点的最大特征值作为输出特征值，将区间内所有尺度下的特征值图进行了融合，经过二值化处理、形态学滤波处理得到了手指静脉纹路。研究结果表明，该算法能够较完整地提取宽度不一的手指静脉纹路，将静脉区域和非静脉区域分开，伪静脉像素点较少；同时，不需要遍历所有像素点的Hessian矩阵求其特征值，手指静脉图像分割速度比未优化静脉纹路提取算法快了0．036 5 S。

英文摘要：

Aiming at extracting the vein p＇atterns in low-quality finger vein images, a finger vein images segmentation algorithm based on Hes-sian matrix was proposed. Hessian matrix was acquired by the convolution of the second derivative of Gaussian filter and image, and the pix-els not belonging to finger vein region were firstly filtered out with the use of the property of matrix trace. Then the Hessian matrix eigenvS.luesof the rest of the pixels were calculated and the pixels not belonging to finger vein region according to the requirement of the eigenvalues werefiltered again. Finally, the maximum of the eigenvalues at each pixel under different scales was chosen as the output, and a finger vein imagewas segmented effectively after image binarization and morphological filtering. The results indicate that the algorithm performs well in separa-ting the vein region from the non-vein region, and the performing is 0. 036 5 s faster than the algorithm not optimized for no need of goingthrough all pixels to calculate the Hessian matrix eigenvalues.