means and tools 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, 1999, p. 1191).
Image fusion forms a subgroup within this definition, with the objective to generate
a single image from multiple image data for the extraction of information of higher
quality (Pohl, 1999). Image fusion is used in many fields such as military, medical
imaging, computer vision, the robotics industry, and remote sensing of the environment. The goals of the fusion process are multifold: to sharpen multispectral images,
to improve geometric corrections, to provide stereo-viewing capabilities for stereophotogrammetry, to enhance certain features not visible in either of the single data sets
alone, to complement data sets for improved classification, to detect changes using
multitemporal data, and to replace defective data (Pohl and van Genderen, 1998). In
this article, we concentrate on the image-sharpening process (iconic fusion) and its
relationship with image scale.
Many publications have focused on how to fuse high-resolution panchromatic
images with lower resolution multispectral data to obtain high-resolution multispectral imagery while retaining the spectral characteristics of the multispectral data
(e.g., Cliche et al., 1985; Welch and Ehlers, 1987; Carper et al., 1990; Chavez et al.,
1991; Wald et al., 1997; Zhang 1999). It was evident that these methods seem to work
well for many applications, especially for single-sensor, single-date fusion. Most
methods, however, exhibited significant color distortions for multitemporal and
multisensoral case studies (Ehlers, 2004; Zhang, 2004).
Over the last few years, a number of improved algorithms have been developed
with the promise to minimize color distortion while maintaining the spatial improvement of the standard data fusion algorithms. One of these fusion techniques is Ehlers
fusion, which was developed for minimizing spectral change in the pan-sharpening
process (Ehlers and Klonus, 2004). In a number of comprehensive comparisons, this
method has tested superior to most of the other pan-sharpening techniques (see, e.g.,
Ling et al., 2007a; Ehlers et al., 2010; Klonus, 2011; Yuhendra et al., 2012). For this
reason, we will use Ehlers fusion as the underlying technique for the following
discussions. The next section presents a short overview of this fusion technique.
2.3 EHLERS FUSION
Ehlers fusion was developed specifically for spectral characteristic-preserving image
merging (Klonus and Ehlers, 2007). It is based on an intensity–hue–saturation (IHS)
transform coupled with a Fourier domain filtering. The principal idea behind spectral
characteristic-preserving image fusion is that the high-resolution image has to sharpen
the multispectral image without adding new gray-level information to its spectral
components. An ideal fusion algorithm would enhance high-frequency changes such
as edges and gray-level discontinuities in an image without altering the multispectral
components in homogeneous regions. To facilitate these demands, two prerequisites
have to be addressed. First, color information and spatial information have to be
separated. Second, the spatial information content has to be manipulated in a way that
EHLERS FUSION
15
obtaining information of greater quality; the exact definition of ‘greater quality’ will
depend upon the application” (Wald, 1999, p. 1191).
Image fusion forms a subgroup within this definition, with the objective to generate
a single image from multiple image data for the extraction of information of higher
quality (Pohl, 1999). Image fusion is used in many fields such as military, medical
imaging, computer vision, the robotics industry, and remote sensing of the environment. The goals of the fusion process are multifold: to sharpen multispectral images,
to improve geometric corrections, to provide stereo-viewing capabilities for stereophotogrammetry, to enhance certain features not visible in either of the single data sets
alone, to complement data sets for improved classification, to detect changes using
multitemporal data, and to replace defective data (Pohl and van Genderen, 1998). In
this article, we concentrate on the image-sharpening process (iconic fusion) and its
relationship with image scale.
Many publications have focused on how to fuse high-resolution panchromatic
images with lower resolution multispectral data to obtain high-resolution multispectral imagery while retaining the spectral characteristics of the multispectral data
(e.g., Cliche et al., 1985; Welch and Ehlers, 1987; Carper et al., 1990; Chavez et al.,
1991; Wald et al., 1997; Zhang 1999). It was evident that these methods seem to work
well for many applications, especially for single-sensor, single-date fusion. Most
methods, however, exhibited significant color distortions for multitemporal and
multisensoral case studies (Ehlers, 2004; Zhang, 2004).
Over the last few years, a number of improved algorithms have been developed
with the promise to minimize color distortion while maintaining the spatial improvement of the standard data fusion algorithms. One of these fusion techniques is Ehlers
fusion, which was developed for minimizing spectral change in the pan-sharpening
process (Ehlers and Klonus, 2004). In a number of comprehensive comparisons, this
method has tested superior to most of the other pan-sharpening techniques (see, e.g.,
Ling et al., 2007a; Ehlers et al., 2010; Klonus, 2011; Yuhendra et al., 2012). For this
reason, we will use Ehlers fusion as the underlying technique for the following
discussions. The next section presents a short overview of this fusion technique.
2.3 EHLERS FUSION
Ehlers fusion was developed specifically for spectral characteristic-preserving image
merging (Klonus and Ehlers, 2007). It is based on an intensity–hue–saturation (IHS)
transform coupled with a Fourier domain filtering. The principal idea behind spectral
characteristic-preserving image fusion is that the high-resolution image has to sharpen
the multispectral image without adding new gray-level information to its spectral
components. An ideal fusion algorithm would enhance high-frequency changes such
as edges and gray-level discontinuities in an image without altering the multispectral
components in homogeneous regions. To facilitate these demands, two prerequisites
have to be addressed. First, color information and spatial information have to be
separated. Second, the spatial information content has to be manipulated in a way that
EHLERS FUSION
15