基于双边滤波算子的医学脊椎去噪  

Medical Spine Denoising Model Based on Bilateral Filter Operator

作　　者：惠宇[1] 武君胜[2] 鱼滨[3] 杜静[4] 李航 

机构地区：[1]西北工业大学计算机学院,西安710072 [2]西北工业大学软件与微电子学院,西安710072 [3]西安电子科技大学计算机学院,西安710072 [4]西北工业大学管理学院,西安710072 [5]法士特集团咸阳精密机械分公司,陕西咸阳712046

出　　处：《光子学报》2017年第7期159-164,共6页Acta Photonica Sinica

基　　金：国家自然科学基金(No.61172147)资助~~

摘　　要：为了有效滤除医学脊椎模型的噪声点,同时更好地保持模型细节,提出了一种基于双边滤波算子的医学脊椎去噪模型.采用双边滤波在多尺度条件下进行脊椎三维模型轮廓线的提取,设计改进自适应扩散系数,以更好的优化控制整个扩散过程.根据图像的离散特征,建立相应的离散迭代方程,使迭代过程离散化,并设计迭代停止准则,当去噪平滑后的图像模型与噪声相关性最小时停止迭代.与经典的向异性扩散模型方法实验结果相比,本方法在解决去噪方面达到了很好的滤波效果,同时也较好地保持了医学图像的边缘细节特征,大大优于传统滤波算法.In order to filter out the noise points of the spine 3D model effectively, a denoising and spreading method for spine 3D model based on bilateral filter operator was proposed. The bilateral filter operator is used to extract the contours of the 3D model, design and improve the adaptive diffusion coefficient, in order to better optimize the control of the entire diffusion process. The discrete iterative equation is established and the iterative stop criterion is designed so that the iteration is stopped when the de-noising effect is optimal. Compared with classical methods such as approximate diffusion model method, the proposed method achieves good filtering effect in solving the denoising method, which is much better than the traditional filtering algorithm.

关 键 词：三维模型 双边滤波 去噪 自适应 特征保持 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]