基于小波变换的遥感图像融合研究——以黄河三角洲地区为例

The Remote Sensing Image Fusion Research Based on Wavelet Transform——Take the Yellow River as an example

针对黄河三角洲地区Landsat TM的多光谱影像和全色影像,提出了一种基于IHS变化和离散小波变换的遥感图像融合方法。该方法在IHS变换的基础上进行直方图均衡化;其次,利用离散小波变换的方法对多光谱图像的亮度分量I和全色影像进行融合;用融合后的图像代替多光谱图像的亮度分量I再进行IHS逆变换得到融合结果。实验证明,融合后的影像不仅保留Landsat TM多光谱图像的光谱特性,而且具有很高的空间细节表现能力。

In this paper, in allusion to the processing of multispectral and panchromatic images of Yellow River Delta area taken by I.andsat TM, the writer has proposed an approach to Fusion of Remote Sensing Images on the basis of IHS chan ges and Discrete Wavelet Transform. Firstly, we carry out IHS change for the multispectral image to achieve three components, including brightness component I, chroma H and saturation degree S. Then extract brightness component I from the multispectral image and implement the histogram equalization. After that, the fusion of brightness component I and panchromatic image will be realized by using fusion method of discrete wavelet transform. Afterwards, take the fused im age to replace brightness component I of the multispectral image. Finally, the fusion result can be achieved by applying in- verse transformation of IHS. It is proved that the spectral character of multispectral image taken by Landsat TM will be retained in the fused image, which also shows good performance of spectral details.