International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
3. RADIOMETRIC DIFFERENCES
The basic premise in using remotely sensed data for a change
detection application is that changes in land cover will result in
changes in radiance values. Moreover, changes in radiance due
to land cover changes must be larger when compared to
radiance changes caused by other factors. These other factors
might include differences in atmospheric conditions, sun angle
and soil moisture. One should expect that these factors will
affect the reliability of change detection algorithms especially
when considering images captured by different sensors that
have varying radiometric and spatial resolutions.
To alleviate the effect of these factors, intensity normalization
is traditionally used as a pre-processing technique to
compensate for possible illumination variations between the
involved images. In this type of pre-processing, the intensity
values in the second image are normalized to have the same
mean and variance values as those in the first image. Assuming
that the involved images are co-registered relative to a common
reference frame, one can proceed by applying image-
differencing methods to create a new image that represents the
changes. The comparison results are based on the assumption
that the differences between the radiometric properties of
corresponding pixels are due to actual changes in the object
space. However, these differences could be the result of other
factors, such as different atmospheric conditions, noise,
different imaging sensors, and/or registration/rectifications
errors. Moreover, the difference image is usually binarized by
thresholding where thresholds are empirically selected. In these
cases, traditional approaches to change detection, which are
based on the differencing of intensity images, fail.
To overcome these problems, derived edges from the registered
images are used as a basis for the proposed change detection
methodology. The utilization of edges is motivated by the fact
that they are invariant with respect to possible radiometric
differences between the images in question.
4. CHANGE DETECTION METHODOLOGY
The proposed method for change detection is as follows:
= Resample the input image into the reference frame
associated with the reference image. The parameters of the
registration transformation function (Section 2) are used in
the resampling process. After resampling, corresponding
pixels are assumed to belong to the same object space
feature.
=» Apply intensity normalization techniques to the images in
question (e.g., to ensure that they have the same intensity
mean and variance values) in order to reduce the radiometric
differences between the involved images. However, this
procedure would not be enough to eliminate radiometric
differences in the involved images.
* Extract edge cells from the resampled images using the
canny edge detector (Canny, 1986). Utilizing the edge
images has two advantages. First, derived edges are robust to
possible radiometric differences between the registered
images (e.g. due to noise and/or different spectral
properties). Also, the edges would correspond to interesting
features (e.g., building boundaries, roads, trails, etc.).
Therefore, comparing edge images will be useful in outlining
the amount of urbanization activities, which is one of the
most important objectives of change detection exercises. The
final output of the edge extraction process will be binary
images in which white pixels refer to edges while black
pixels refer to non-edges.
* Apply the majority filter (also known as the mode filter) to
the edge images. This filter is applied to binary images
where a window is centered at each pixel and the value of
this pixel is changed or maintained according to the majority
of the pixels within this window (Lillesand and Kiefer 2000).
In the proposed methodology for change detection, the
majority filer has been implemented for the following
reasons:
e To compensate for small registration errors (in the
order of few pixels).
e To balance the effect of varying edge densities in the
registered images especially when dealing with multi-
source images.
e To fill small gaps within an area with numerous edges
(Figure 2-a, highlighted by solid circles), and smooth
object boundaries without expanding and/or shrinking
the objects (Lillesand and Kiefer 2000).
*e To eliminate isolated edges (Figure 2-b, highlighted by
dotted circles).
(b) -Bef
(b)— After
Figure 2. Majority filter: filling gaps among dense edges (a),
448
removing isolated edges (b)
As a result, filtered images will highlight areas with
interesting features since they lead to a dense distribution of
edge cells.
= Subtract filtered images (pixel-by-pixel) in order to highlight
areas of change.
* Apply a majority filter to the difference image to eliminate
small areas (since changes/no-changes are expected to be
locally spread — i.e., they are not isolated).
5. EXPERIMENTAL RESULTS
Experiments have been conducted using multi-source, multi-
resolution, and multi-temporal imageries over the city of
Calgary, Alberta to illustrate the fcasibility of the suggested
methodology. The experiments incorporated a 1374 rows by
1274 columns aerial photo (5.0m resolution) captured in 1956,
1374 rows by 1274 columns aerial photo (3.5m resolution)
captured in 1972, 2000 rows by 2000 columns ortho-image
Intern
(5.0m
1999,
resolu
differ
segme
exam]
ortho-
digitiz
have |
Figure
A clo
corres
referei
incorp
involv
the co
Thee
and th
are li:
which
show :
few p
validit
transfc
A fü ]
ó;
oll
aj
£5
a,
| b, (p
[hu
b,
Table
datase!
In add
betwec
Figure
digitiz
ortho-
city is
acrial
image
in the
feature
each ©
within
the ch
during
