不同遥感影像融合方法效果的定量评价研究

Quantitative Assessment of Image Fusion Methods

随着遥感影像融合技术研究的发展,提出了许多遥感影像融合的方法。在实际应用中,为了正确和有效地利用融合方法,定量评价不同融合方法的效果是非常必要的。本文对小波法(WT)、Gram_Schimdt法(GS)、合成变量系数法(SVR)以及基于平滑滤波的亮度变换法(SFIM)四种融合方法,在中等分辨率、高分辨率两种情况下,从影像融合的光谱信息保真度和空间信息融入度两方面进行了定量评价。结果表明:在空间信息融入方面,无论是中等分辨率影像融合,还是高分辨率影像融合,SVR法效果最佳,GS法、SFIM法和WT分别次之;在光谱信息保真方面,在中等、高分辨率影像融合中的效果有所不同,其中在中等分辨率影像融合中,GS法效果最佳,SVR法、SFIM法、WT法分别次之,而在高分辨率影像融合中,GS法效果最佳,SVR法、WT法、SFIM法分别次之。

With the development of the image fusion techniques, many methods were put forward. In order to make full use of these methods practically, it is necessary to assess their effectiveness quantitatively. In this paper, in case of medium resolution image（Landsat 7 ETM＋ image） and high resolution image（QuickBird image）, the effectiveness of four methods, WT（Wavelet trarsform）,SFIM（Smoothing Filter-based Intensity Modulation）,GS（Gram Schimdt transform）and SVR（Synthetic Variable Ratio） ,were assessed for spectral information fidelity and spatial information syncretizing. The results showed that in terms of spatial information syncretizing,SVR transform was the best, followed by GS, SFIM and WT transforms in both medium and high resolution image fusion,but in terms of spectral information fidelity,it was different,that GS transform was the best, fol- lowed by SVR,SFIM and WT transforms in medium resolution image fusion,and GS transform was the best,followed by SVR, WT and SFIM transforms in high resolution image fusion.