International Archives of Photogrammetry and Remote Sensing, Vol. 32, Part 7-4-3 W6, Valladolid, Spain, 3-4 June, 1999 
TOOLS AND METHODS FOR FUSION OF IMAGES OF DIFFERENT SPATIAL RESOLUTION 
C. Pohl 
International Institute for Aerospace Survey and Earth Sciences (ITC), P.O. Box 6, 7500 AA Enschede, The Netherlands, 
pohl@itc.nl 
KEYWORDS: Image fusion, pixel level, resolution merge, spatial resolution, fusion techniques. 
The usefulness of remote sensing data, in particular of images from Earth observation satellites, largely depends on their spectral, 
spatial and temporal resolution. Each system has its specific characteristics providing different types of parameters on the observed 
objects. Focussing on operational and most commonly used commercial remote sensing satellite sensors, this paper describes how 
image fusion techniques can help increase the usefulness of the data acquired. There are plenty of possibilities of combining images 
from different satellite sensors. This paper concentrates on the existing techniques that preserve spectral characteristics, while 
increasing the spatial resolution. A very common example is the fusion of SPOT XS with PAN data to produce multispectral (3- 
band) imagery with 10 m ground resolution. These techniques are also referred to as image sharpening techniques. A distinction has 
to be made between the pure visual enhancement (superimposition) and real interpolation of data to achieve higher resolution (e.g. 
wavelets). In total, the paper describes a number of fusion techniques, such as RGB colour composites, Intensity Hue Saturation 
(IHS) transformation, arithmetic combinations (e.g. Brovey transform), Principal Component Analysis, wavelets (e.g. ARSIS 
method) and Regression Variable Substitution (RVS), in terms of concepts, algorithms, processing, achievements and applications. It 
is mentioned in which way the results of various techniques are influenced by using different pre-processing steps as well as 
modifications of the involved parameters. All techniques are discussed and illustrated using examples of applications in the various 
fields that are part of ITC’s educational programme and consulting projects. 
ABSTRACT 
1. INTRODUCTION 
higher resolution (e.g. wavelets); the latter being proposed 
amongst others by Mangolini (1994) and Ranchin et al. (1996). 
According to the EARSeL 1 Special Interest Group on Data 
Fusion (Data Fusion SIG) data fusion is defined as "... a formal 
framework in which means and tools are expressed for the 
alliance of data originating from different sources. It aims at 
obtaining information of greater quality; the exact definition of 
greater quality will depend upon the application" (Wald, 1998). 
Image fusion forms a subgroup within this definition and aims 
at the generation of a single image from multiple image data for 
the extraction of information of higher quality. Having that in 
mind, the achievement of high spatial resolution while 
maintaining the provided spectral resolution falls exactly into 
this framework. 
Using appropriate fusion techniques high spatial resolution 
panchromatic imagery can be combined with multispectral 
imagery of lower resolution. In this way, the spectral resolution 
may be preserved, while a higher spatial resolution is 
incorporated, which represents the information content of the 
images in much more detail (Franklin and Blodgett, 1993; 
Pellemans et al., 1993). A special case is the fusion of channels 
from a single sensor for resolution enhancement, e.g. TM data. 
The lower resolution thermal channel can be enhanced using the 
higher resolution spectral channels (Moran, 1990). Other 
approaches increase the spatial resolution of the output channel 
using a windowing technique on the six multispectral bands of 
TM (Sharpe and Kerr, 1991). The fusion of SAR/VIR does not 
only result in the combination of disparate data but may also be 
used to spatially enhance the imagery involved (Welch, 1984). 
Geometric accuracy and increase of scales using fusion 
techniques is of concern to mapping and updating (Jutz, 1988; 
Chiesa and Tyler, 1990; Pohl, 1996). 
2. RESOLUTION MERGE 
The concept of combining images with complementary 
information opens a broad field of applications. There is a vast 
variety of techniques to combine images from different sensors. 
However, this paper focuses on image fusion techniques that 
preserve spectral characteristics whilst increasing spatial 
resolution to provide images of greater quality. A very common 
example is the fusion of SPOT XS with PAN data to produce 
multispectral (3-band) imagery with 10 m ground resolution. 
These techniques are also referred to as image sharpening 
techniques and often called resolution merge. A distinction has 
to be made between the pure visual enhancement 
(superimposition) and real interpolation of data to achieve 
3. IMAGE FUSION TECHNIQUES 
European Association of Remote Sensing Laboratories. 
Image fusion for spatial resolution enhancement is performed at 
pixel level as one of the three fusion levels defined by Pohl and 
van Genderen (1998). It requires the accurate co-registration of