摘要
基于医学影像MRI和CT图像互相联系、互相补充的特点,为了更好地实现不同模态医学图像间的融合,提出一种改进的基于区域特征的医学图像融合方法。原理是根据图像小波变换系数的一系列参数如区域均值、方差、协方差等构造加权因子和匹配度,对代表图像细节信息的高频子图进行区域融合;然后对低频部分采取选择极大绝对值的原则,对图像进行融合。最后经过小波重构得到输出图像。经过主观和客观评价对实验结果进行综合评价,此方法处理过的图像能得到良好的视觉效果和理想的指标,时间上也减少了0.25s,适用于实时系统的应用。
Based on the features that medical image MRI and CT have interaction with each other, for the better effect of medical image fusion, an improved fusion algorithm based on provincial characteristics was put forward. The principle is to construct weighted factor and matching degree with some related parame- ters to compound the area of high frequency which presents the detailed information. To the area of low frequency, we select the principle of maximum absolute value. Finally we get the fusion image from wave- let reconfiguration. By the estimate of subjectivity and objective, the method reduced running time by 0.25 s that it is applied and could export excellent visual effect and good parameters.
出处
《青岛大学学报（自然科学版）》
CAS
2011年第2期33-38,共6页
Journal of Qingdao University(Natural Science Edition)
关键词
医学图像融合
区域特征
小波变换
medical image fusion
provincial characteristies
wavelet transform
