国产高分辨率遥感卫星融合方法比较

Comparison of Fusion Methods in Domestic High Resolution Remote Sensing

遥感影像融合技术不仅提高了遥感影像的空间分辨率,还保留了原始遥感影像的光谱信息,因此,融合在遥感影像处理领域应用较为广泛。本文针对资源三号和高分二号影像的特点,选择主成分变换、乘积变换、G-S融合算法、HSV融合、Pansharp融合、Brovery变换融合方法进行影像融合试验,并采用主客观方法进行融合质量评价。实验结果表明:对于资源三号影像,Pansharp融合方法效果最优,融合后影像适应性也最强;G-S融合算法和HSV融合方法比Pansharp融合方法稍差,主成分变换和乘积变换最差。对于高分二号影像,Pansharp融合效果最好,Brovey变换融合次之,乘积算法最差。因此,在实际应用中,应根据卫星影像的不同选择适当的融合算法,实现更好的应用。

The remote sensing image fusion technology not only improves the spatial resolution of remote sensing images,but also preserves the spectral information of the original remote sensing images.Therefore,the fusion is widely used in the field of remote sensing image processing.In this paper,according to the features of ZY-3 and GF-2 images,we choose the principal component transform,product transform,G-S fusion algorithm,HSV fusion,Pansharp fusion,and Brovery transform fusion method to carry out image fusion experiments,and use subjective and objective methods for fusion quality evaluation.The experimental results show that the Pansharp fusion method has the best effect for ZY-3 image,and the fusion image has the strongest adaptability.The G-S fusion algorithm and the HSV fusion are slightly inferior to the Pansharp fusion,and the principal component transformation and the product transformation are the worst.Pansharp fusion works best for GF-2 image,followed by Brovey transforms and worst on product algorithms.Therefore,in practical applications,appropriate fusion algorithms should be selected based on the different satellite images to achieve better applications.