nbul 2004 
both keep 
ough grid 
ility. 
ese three 
classified 
od are the 
itive. The 
nd-pepper 
band/ the 
esholding 
structured 
for each 
shold and 
mance of 
ess of the 
oise than 
thods, the 
h doesn't 
the worst 
polarized 
filter [3]. 
rt of the 
a multi- 
' mean of 
lying the 
ulti-band 
nents are 
ample of 
1 to the 
s are the 
omplete” 
g this set 
ich layer 
terested. 
ch layer 
orm the 
more, in 
ch as in 
ed in the 
used in 
ch layer 
igure 6, 
shallow 
cted one 
ws the 
red are 
taken in 
-of-land 
cept the 
So they 
are the 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004 
  
4. TOPOGRAPHIC CHANGE DETECTION 
4.1 Change Detection Based on Region Classification 
Topographic change detection is studying changes on the 
surface of the earth. Satellite images are used to perform 
topographic detection at very high accuracy. In this paper we 
present a topographic change detection method that applies the 
automatic update algorithm presented in [1]. Namely, for a two 
level classification problem we consider an image I — S, * S., 
where S;, i=1,2 are compact regions of the image represented by 
contours. The contours enclose pixels that correspond to the 
same region. When a change occurs, two groups of pixels are 
changing region. Those that move from region S, to region S; 
are named as "additions" (A), while the others that change from 
region S5 to region S, are called "deletions" (D). The total 
change C in the image ] is expressed as the summation of 
additions and deletions, C=A+D. Namely, 
79 eU +52 
J Su + gun 
= se p guo f 59 
= se > sen ns® 
where the subscript “i” and “i-1” are the time index, which 
represent the current and previous time, respectively. The 
advantage of this method is details of the changes are provided. 
In a log of applications, we are not only interested in where the 
changes happen, but also how are the changes. 
In the following, we will extend this concept to multiple region 
cases and automatic update of information. In a distributed 
processing system this mechanism may be programmed to keep 
updates of changes of classification regions or other features 
over time. 
For the images have multiple level classification, we are 
interested in the changes in each region, ie. addition and 
deletion. Assume we have M interested regions, which are 
presented in M “layers”, L,L,,A , L,,, Where the region-of- 
interest in the layer L, is denoted as g . The pixels in g have 
values “1” and all the other pixels are set to zeros. The basic 
idea is comparing the pairs of layers of different times one by 
one. Namely, find the addition and deletion for each L. For 
cach pair of layers, the region-of-interest p is exactly the S, in 
our previous discussion, and the other part having zero values is 
the S... In this way, if we use Z‘ to denote the Ath layer at time 
I, the common region of interest will be 
(7-1) GN G) 2 
RU ERU mM oL (2) 
where the operator *O" represents the element-by-element 
* 
multiplication of two matrices, and of ) " represents a region 
which is composed of the pixels whose values are ones in **". 
In this way, the addition of &, therefore will be 
i / n i- I) (3) 
A cus S 
K 
and the deletions is 
i- i—1} i) (4) 
D gu IL. 
The total change for the À th region will be 
c = AT > 9n = un 4 js s 215 oii. (5) 
However, if we perform frame-to-frame subtraction, we will 
obtain 
REY A RU = auem ol’), (6) 
and we have 
a“ = en (7) 
From the above we can see in a two level classification 
problem the total change may be expressed through the 
absolute value of a frame-to-frame differencing. In practice 
we are interested in more details of the changes such as 
additions and deletions. Our proposed formulations give 
these details. 
Apply the above procedures to each pair of layers. Step by 
step the addition and deletion for every class will be detected 
sequentially. 
1 
4.2 Change Detection Based on Pixel Level 
Characteristics 
The ability to preserve the pixel characteristics from frame to 
frame when change detection is performed is essential if 
multiple classification inferences are derived from the 
changes. In this case image classification process is carried 
out on the change detection results. Two methods have been 
studied for change detection on images with multiple 
classification regions, i.e. the principal component analysis 
and the wavelet method. 
4.3 Matched filtering and change detection. 
Change detection may be applied using matched filters. 
Matched filters tend to suppress clutter and emphasize the 
changes of interest. When matched fiiters are applied the 
change detection performance increases. Matched filtering 
for change detection is normally applied to multispectra 
and/or multipolarized images [3]. [5]-[7]. 
S. EXAMPLE 
Let's consider the two images in Figure 4. Their layers are 
presented in Figure 9. We apply the proposed multi-level 
change detection method to the pair of layers {(a), (d)}, {(b), 
(e)} and {(c), (D}, respectively. The result is displayed in 
Figure 10, where the red regions represent deletions, green 
ones stand for additions, and the yellow means no change 
happens. Figure 10 clearly gives the details of change in each 
region. lt is easy to find from Figure 10 (c), because of the 
flooding in May, some regions of shallow water and land in 
the image of August become deep water (the red region in 
(c)). For the same reason, in (a), the green regions are the 
parts that are changed from shallow and deep water in May to 
land in August. Using this method avoids need for strict 
radiometric calibration. We can choose the appropriate 
classification scheme. The most important is it designates the 
types of changes occurring. It is simple, reliable and 
effective.