摘要
多小波是小波理论的扩展,在图像处理方面具有单小波所不具有的优点.它能够为图像提供一种比小波多分辨率分析更加精确的分析方法.在研究了多小波变换域上同一尺度多个子带间相关性、子带内相邻系数的相关性以及能量的低聚性的特性后,提出了一种基于离散多小波变换域特征的融合方法,并将不同模态的医学脑部CT图像和MR图像利用此方法进行融合,相比于传统小波域内的图像融合方法.该方法不仅能够完好地显示源图像各自的信息,很好地将源图像的细节融合在一起,而且得到的融合图像具有更良好的视觉效果和更优的量化指标,体现出更好的融合效果.
Multiwavelet is an extension from wavelet theory,and has several particular advantages in comparison with scalar wavelets on image processing.Multiwavelet analysis can offer a more precise way for image analysis than wavelet multi-resolution analysis.In this paper,we research the characteristics of the strong correlativity among sub-images and the correlativity among neighboring coefficient of sub-image and the energy converging to the lowest sub-images of multiavelet transform and give a fusion method based on fields features of discrete multiavelet transform.In the experiment,we have obtained the fused picture of multimodality medical image of brain CT and MR by applying this image fusion methed.Comparing with fused methed by means of wavelet transform and evaluating them in way of objective and subjective performance,we can draw the conclusion that this method can fuse details of input images successfully,and it can perfectly disply information of the each input image ,therefore using this image fusion method can get more satisfactory result than using others.
出处
《电脑知识与技术（过刊）》
2007年第18期1709-1711,共3页
Computer Knowledge and Technology
基金
国家自然科学主任基金(60642009)
关键词
多模态医学图像
图像融合
离散多小波变换
Multimodality Medical Image
Image Fusion
Discrete Multiwavelet Transform
