针对像素级遥感图像融合处理数据量大、实时性差, 融合结果不能直观表述目标特征的问题, 提出一种边缘特征加强的多光谱与全色图像快速融合方法。为提高算法的运行速度, 采用提升小波与快速IHS变换相结合的算法生成像素级融合图像。同时, 利用数学形态学方法提取全色图像的边缘, 与像素级融合图像进行再次融合, 加强了最终融合图像的边缘信息。实验结果表明, 该算法不仅融合效果好, 而且实时性强。

Considering that the process of pixel-level remote sensing fusion with large amount of data has poor real-time performance and the fusion result can not describe target feature intuitively we proposed a fast fusion method for panchromatic and multi-spectral images with enhanced edge feature. In order to improve the running speed a pixel-level fusion was generated by the algorithm combining lifting wavelet transform with fast IHS transform. Meanwhile the edges of panchromatic images were extracted by mathematic morphology which were then fused again with the pixel-level fusion image to generate the final fusion image with enhanced edge information. The experimental result indicates that the proposed algorithm can not only obtain a good fusion result but also has a well real-time performance.