不同融合方法对遥感影像融合质量的影响

Influence of Different Fusion Methods on the Fusion Quality of Remote Sensing Image

遥感影像融合是影像处理的重要环节,是多源遥感影像数据处理的一种主要手段,其中融合方法的选择对融合结果来说至关重要。为探讨不同融合方法对遥感影像融合质量的影响,采用Pansharp融合法、HIS融合法、小波融合法和HPF融合法,对不同国产卫星遥感影像进行融合,并从光谱保真度和影像信息量两方面,选用定量评价常用的均值、信息熵、相关系数、标准差4个指标对融合后的影像进行质量评价。结果表明:从综合角度来看,Pansharp融合法得到的融合影像的信息量增加量效果最明显,HIS融合法对影像光谱保真度相对其他方法来说较好;在主观视觉上,Pansharp融合法得到的影像,目视效果好,影像清晰度较高,色彩更真实。

Remote sensing image fusion is an important link in sensing image processing and a main means of multi-source remote sensing image data processing, and the selection of fusion method is very important for the fusion result. In order to explore the influence of different fusion methods on the fusion quality of remote sensing images, this paper adopts Pansharp fusion method, HIS fusion method, wavelet fusion method and HPF fusion method for image fusion of different domestic satellite remote sensing images. And from the two aspects of spectral fidelity and information content of the images, the mean value, information entropy, correlation coefficient and standard deviation, which are commonly used in quantitative evaluation methods, are used to evaluate the quality of the fused images.The results show that from a comprehensive perspective, Pansharp fusion method has the most obvious effect in increasing the information content of the fused image, while HIS fusion method has relatively good spectral fidelity of the image;In terms of subjective vision, the image obtained by Pansharp fusion method has good visual effect, higher image definition and more realistic color.