基于球尺度的医学图像矢量滤波算法研究  

Research on Medical Image Filtering Algorithm Based on Vector Ball Scale

作　　者：陈利军[1] 郭红艳[2] 

机构地区：[1]河南经贸职业学院,郑州450018 [2]郑州信息科技职业学院,郑州450046

出　　处：《激光杂志》2016年第1期135-137,共3页Laser Journal

基　　金：河南省高等学校青年骨干教师资助计划项目(教高〔2013〕984号);2015年度河南省软科学研究计划项目(152400410185;152400410597)

摘　　要：医学图像中常常带有高斯脉冲,导致医学图像中的一些重要信息模糊不清晰,从而严重影响了医学图像视觉效果。在医学图像滤波中,既要对图像中的噪声有效去除又要保证医学图像滤波后尽可能的保证图像的边缘和细节等重要特征不被破坏。为此本文提出了一种基于球尺度的医学图像矢量滤波算法,在传统的矢量中值滤波器(VMF)基础上引入了球尺度概念,从而得到了基于球尺度的矢量中值滤波器(BSVMF)。新的滤波方法在图像边界进行小平滑处理,在图像内部区域进行大平滑处理,从而实现了对滤波过程的自适应控制。实验结果说明,本文提出的新滤波方法能够在消除噪音污染的同时,也能够尽可能多的保留图像边缘或尖角等重要信息。Medical image with Gaussian pulse used in, causes some important information in medical image not clear, which seriously affects the Visual effect of medical images. In medical image filtering, not only effectiuely remove the image noise, but also ensure that the edges of the image and details and other important characteristics are not dam- aged. This paper proposes a vector transform algorithm for medical image based on ball scale, introduce the ball scales concept into the traditional vector median filter（VMF） then obtain based on the scaling vector median filter（BSVMF）. The new filtering method smoothing the edges within a narrtow range, smoothing the internal area of the image, thus a- chieving the Adaptive control of the filtering process. The experimental results show that the proposed new filtering method can eliminate noise pollution, are able to retain as much as possible image edges or sharp corners, and other im- portant information.

关 键 词：医学图像 高斯脉冲 球尺度 矢量中值滤波器 自适应 

分 类 号：TN911.73[电子电信—通信与信息系统]