一种基于小波变换的医学断层图像插值方法  

An Improved Interpolation of Medical Images

作　　者：王伟静[1] 陈家新[1] 

机构地区：[1]河南科技大学电子信息工程学院,河南洛阳471003

出　　处：《河南科技大学学报（自然科学版）》2006年第3期29-31,36,共4页Journal of Henan University of Science And Technology:Natural Science

基　　金：河南省自然科学基金项目(0411010200)

摘　　要：图像插值技术是图像三维重建的关键技术。为了满足灰度和形状同时插值的要求,提出了一种断层图像插值方法,其技术特点是通过小波变换得到图像边缘小波系数强度插值,然后采用一种寻找匹配点的方法对由小波变换得到的低频子图进行灰度插值,最后将所得到的新的高频小波子图和低频子图形成新的插值序列,并利用小波逆变换得到新的插值图像。实验结果表明,所得到的插值图像不仅在形状上更接近于自然状态,在灰度也满足了医学图像插值的要求。Interpolation of medical image is a key technology in 3D reconstruction.In order to interpolate gray value and shapes at the same time,a new interpolation method is introduced.Firstly,the original images are decomposed with wavelet transform to obtain the wavelet coefficients of those edge points,which are then interpolated to get three new higher frequency sub-images.Secondly,the two lower frequency wavelet sub-images are dealt with,with matching points between two corresponding sub-images being sought and their grey value interpolated.Finally,the reverse wavelet transform is adopted to get the final interpolation image.Experiment results show the final image is better in both grey-level and image contour and it can meet the requirements of medical image interpolation.

关 键 词：图像插值 小波变换 匹配点 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]