压缩感知在采样的同时完成信号的压缩,为解决医学图像融合过程中时间复杂度高、传输数据量大的问题提供了新方法.本文围绕压缩感知在医学图像像素级融合做了5个方面的工作:第一,给出了基于压缩感知的医学图像融合框架;第二,讨论了基于贝叶斯、贪婪迭代、凸松弛等四类重构算法;第三,梳理出医学图像像素级融合的七类方法;第四总结出基于压缩感知的四种医学图像融合路径;第五,指出了目前研究的难点和应用前景.%Compressed sensing (CS) is a new compression sampling technology,which can reduce the sampling data,storage and transmission.CS provides a new way to resolve high time complexity,large transmission data in medical image fusion.Five aspects of pixel-level fusion in medical image are discussed in this paper.Firstly,a framework of medical image fusion based on compressed sensing is putted forward by this paper.Secondly,four kinds of reconstruction algorithm,such as Bayesian,convex relaxation,greedy iterative,are discussed comprehensively.Thirdly,seven kinds of medical image fusion in pixel-level fusion are summarized,Fourth,four paths of medical image fusion based on Compressed sensing are summarized.Finally,The difficulties and application prospects of the research are pointed out.
展开▼