摘要

The enhancement of peripheral nerve CT images is the foundation of 3D peripheral nerve reconstruction. Peripheral nerve segmentation and reconstruction are restricted by the poor image. In this paper, we propose a new multi-scale peripheral nerve enhancement method from Hessian matrix. Peripheral nerve images are filtered using Gaussian filter, after being normalized. Then, an enhancement filter function is presented to enhance peripheral nerve CT images through eigenvalues analysis of Hessian matrix for Gaussian circle model. Experimental result shows that the algorithm can not only enhance circular region but also suppress other interference regions, which will be beneficial to segmentation and reconstruction.

全文