一种基于区域和Contourlet变换的多聚焦图像融合新方法

本文提出了一种基于区域和contourlet变换相结合的图像融合新方法。该方法首先对源图像进行contourlet变换，得到高频和低频图像；接着根据各尺度上高频图像的大小来调整源图像尺寸进行分割，得到多尺度分割图像；然后采用区域高频系数绝对值之和作为区域活动度来指导系数融合；最后进行contourlet逆变换得到融合图像。采用均方根误差、信噪比、边缘融合质量指标和加权融合质量指标4种准则来评价融合算法的性能。实验结果表明，本文方法不仅在客观评价指标上优于小波变换法，而且从主观评价上来看，本文的方法得到的融合图像更加的清晰。

A new method for muti-focus image fusion based on contourlet transform and region

A New method for multi-focus image fusion based on contourlet transform and region is proposed.Firstly,contourlet transform is performed on source images to get high-frequency and low-frequency images.The source images are resized depending on the size of high-frequency images at each scale and then they are segmented into different regions.The sum of absolute value of high frequency coefficients in a region is considered as the region activity to conduct coefficient fusion.At the end,the inverse contourlet transform is utilized to obtain fused image.The performance of the image fusion method is evaluated using four criteria including root mean square error,signal-to-noise ratio,edge-dependent fusion quality index and weighted fusion quality index.The results of experiments imply that the new method not only get higher object assessment index score but also get a more clearly fusion image in perceptual than wavelet transform method.