Remote sensing image fusion using the curvelet transform
Information Fusion
Multispectral images fusion by a joint multidirectional and multiresolution representation
International Journal of Remote Sensing
Modern computational techniques for environmental data; application to the global ozone layer
ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part III
| Hi-index | 0.00 |
Fusion of panchromatic (PAN) and multispectral (MS) images is one of the most promising issues in remote sensing. PAN modulation fusion methods are usually based on an assumption that a ratio of two different-resolution versions of an MS band is equal to a ratio of two different-resolution versions of a PAN image. In such fusion methods, image haze is rarely taken into account, and it may produce serious spectral distortion in synthetic images. In this paper, assuming that the previous ratio relationship only holds for haze-free images, two relevant improvement schemes are proposed to better express the ratio relationship of haze-included images. In a test on a spatially degraded IKONOS dataset, the first scheme synthesizes an image with minimum spectral distortion, and the second modifies several current PAN modulation fusion methods and generates high-quality synthetic products. The experiment results confirm that image haze can seriously impact the quality of fused images obtained by using PAN modulation fusion methods, and it should be taken into account in relevant image fusion.