一种改进的RS图像融合方法用于土地利用调查

常用的遥感图像融合的方法如IHS变换法在实施图像融合时,会产生色彩扭曲现象,本文探讨了一种新的光谱保持型的高通滤波融合算法(HPFF融合法)。HPFF融合法先对参与融合的全色波段进行高通滤波,尔后将滤波后的全色波段替换多光谱经IHS正变换后的强度分量,再进行IHS逆变换便得到融合图像。HPFF融合法的光谱保持性能优于IHS变换法,HPFF融合后的图像的色彩与TM影像相近,HPFF融合后图像的分类精度高于IHS融合后的图像,HPFF融合法是一种能较好地保持光谱特性的融合方法。

An Improved Method of RS Image Fusion Applied to Investigation of Land Use

These fusion methods such as IHS transform could merged two optical image data of different resolutions the high spatial resolution panchromatic image but multi-spectral image. However, this fusion method required the spectral response range of the high spatial resolution panchromatic image was the same equal or approximate to the spectral response range of the multi-spectral image. This paper put forward a kind of new fusion approaching called high-pass filter fusion (HPFF) that could merge two optical image data of different spectral characteristic. This paper proposed its algorithm, firstly filter on the panchromatic image with high pass filter, then fuse the remote sensing data applying IHS transformation. The color of the HPFF image is approximate to color of the low spatial resolution multi-spectral image. The HPFF method is more excellent at preserving spectrum than the IHS transformation, and the classification accuracy of the HPFF image is more 5.69% higher than the IHS image.