高原山区星载合成孔径雷达数据与多光谱数据的图像融合探究——以贵州省毕节市为例

为了推动雷达数据在贵州高原山区的应用,针对高原山区的气候与地形特点,基于雷达数据的优势,本文对此类高原山区应用星载雷达数据进行初步探讨。利用TM影像4,3,2波段和雷达SAR数据,通过不同融合方法进行融合,应用标准差、信息熵、相关系数等参数对融合效果进行分析。结果表明:几种方法得到的图像都在不同程度上增强了多光谱图像的光谱信息,同时也增加了雷达数据的纹理信息。其中PC变换能够最好地保持TM数据的多光谱特性,对于植被信息最为突出;HSV变换很好地保留了SAR数据的纹理信息,有利于居民点、建筑物等目标的识别;GS变换增加了图像的可识别度,但降低了整体的清晰程度;Brovey变换效果最差,图像失真较大。

Image Fusion of Spaceborne SAR Data and Multi Spectral Data for Mountainous Plateau: A Case Study on Bijie City,Guizhou Province, China

According to the characteristics of climate and terrain of the plateau area in Guizhou Province, China, application of space borne radar data in such a plateau area based on the advantage of radar data was discussed in this paper to promote the application of radar data in Guizhou Province. By using Band 4,3, and 2 of TM image data and SAR data,effect of image fusion by the parameters including standard deviation, information entropy, correlation coefficient were studied. Results show that several methods of image fusion enhance the spectral information of multi spectral images, but also increase the texture information of radar data in different extents. The PCA transformation can best maintain multi spectral characteristics of TM data, especially for vegetation information. The HSV transformation retains the texture information as well as SAR data, and it will help to identify settlements, buildings and other targets. The GS transformation improves the recognition of images but reduces the overall degree of clarity. The Brovey transformation images have the worse effect with larger image distortion.