中文摘要：

将不同模态的医学图像进行合理融合,能够极大地丰富图像的信息,提高临床医学的诊断效果.本文结合压缩感知理论,采用Contourlet变换对原始图像进行分解,得到低频子带部分和高频子带部分;对低频子带使用平均加权的方式进行融合,对含有较多噪声的高频子带使用邻域能量加权的融合规则.通过对医学图像的仿真实验,该算法可以增加多模态医学图像互补信息,并能较好地提高医学图像融合的清晰度.比较融合评价指标,证实该算法无论在主观视觉还是客观评价指标上都有较好的效果.

英文摘要：

Multi-model medical image fusion can greatly enrich the image information and improve the diagnosis of clinical medicine. In this paper, along with the theory of compressed sensing, we take Contourlet transform for the decomposition of original images to get the low frequency part and high frequency part; the low frequency sub-bands is fused by using a weighted average approach, and the high-frequency sub-bands with high noise content is fused by using weighted local energy fusion rule. Through medical image simulation, the algorithm can increase the multi-modal medical image complementary information, and can improve the clarity of medical images fusion. The fusion evaluations confirm that the algorithm has good results in terms of subjective and objective evaluation.