提出的图像信息融合方法用于解决传统方法在空间分辨率等方面的不足.原始图像经过小波变换,分解成子图像,再进行分块处理,高、低分辨率图像根据全局方差准则分别计算融合的权值系数,进行图像融合重建,然后,对图像采用全局法进行信息融合.实验表明,方法切实可行,能使低分辨率图像接近于高分辨率图像.

A method of information fusion for collectivity multi-sensor remote sensing image is presented based on wavelet transform. The disadvantages of traditional information fusion method on spatial resolution is improved. The original image is transformed into sub-image with wavelet transform and each sub-image is divided into sub-blocks. The weighting value coeffients of high resolution images and low resolution images are computed according to collectivity square error rule, and then the high resolution image and low resolution image are fused and reconstructed. Afterwords, information fusion is made on both of the images with collectivity square error rule. Experimental results show that the method is workable to make low resolution image approaching high resolution image.