International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
e Discontinuities appear along the boundaries between some of 
the resampled patches in Figure 5(d) (highlighted by hollow 
circles). These discontinuities are attributed to real changes 
in the object space between the epochs of capture of the 
involved scenes (the Aerial photo has been captured forty 
three years earlier than the ortho photo scene). This is 
significant in change detection application where accurate 
image registration is a prerequisite for accurate and reliable 
  
validity and image map registration for change detection and 
updating applications. 
8. REFERENCES 
Boardman, D., I. Dowman, A. Chamberlain, D. Fritsch, and W. 
Newton, 1996. An automated image registration system for 
SPOT data. /nternational Archives of Photogrammetry and 
Remote Sensing, 31 (4): 128—133. 
change detection results. ; Kl 
Dare, P., and I. Dowman, 2001. An improved model for 
automatic feature-based registration of SAR and SPOT images. 
ISPRS Journal of Photogrammetry and Remote Sensing, 56 AI 
(2001): 13-28. 
: : : : ; GI 
Fonseca, L., and Manjunath, B., 1996. Registration techniques 
for multisensor remotely sensed imagery. Photogrammetric ps 
Engineering and Remote Sensing, 62(9):1049—1056. ad 
gr: 
Habib, A., and D. Kelley, 2001a. Single photo resection using to 
the modified Hough transform. Photogrammetric Engineering ga 
and Remote Sensing, 67(8): 909-914. tes 
: | su] 
Habib, A., and D. Kelley, 2001b. Automatic relative orientation Wi 
of large scale imagery over urban areas using modified iterated or 
Hough transform. International Journal of Photogrammetry b 
and Remote Sensing, 56 (2001): 29-41. e 
Habib, A., and M. Morgan, 2002. Epipolar image resampling 
from push-broom imagery: investigation and preliminary 1 
implementation. Korean Electronics and Telecommunications 
Research Institute (ETRI), Daejeon, Korea, 107 p. Ki 
: 1 
Habib, A., Y., Lee, and M. Morgan, 2001. Surf: ze matching CO! 
and change detection using modified iterative !.ough transform CO! 
Figure 5. Multi-image mosaic for robust parameter estimation. Photogrammetric + wd, 17 me 
(98): 303-315. cal 
7. CONCLUSION AND REMARKS : as 
Hsieh J., H. Liao, K. Fan, M. Ko, and Y. Hung, 1997. Image the 
This paper addressed the key issues of an efficient semi- registration using a new edge-based approach. Computer Vision rec 
automatic registration methodology that can handle multi- and Image Understanding, 67 (2): 112-130. Sid 
temporal multi-sensor images. The presented approach uses 
; ; ; i Kubik, K., 1991. Relative and absolute orientation based on clo 
linear features (straight-line segments) as the registration js : : e y ass di 
. “ie . . - 1 X y ^ ^P ^ ofpywu 0 o 
primitives since they can be reliably extracted from the linear features. /SPRS Journal of Photogrammetry and Remote A 
A . : . . 21 CT (13: Q. 
reference and input images. Then, the registration Sensing, 46(1): 199-204. e 
transformation function is analyzed to determine the ; ; 
: wail. oR tel ; Seedahmed, G., and L. Martucci, 2002. Automated image en 
mathematical relationship between conjugate primitives in the istrati éricaily | Ant t : | 
v : regis sing eg etrice jar ter spac 
scenes to be registered. It has been established that affine Wid ton on IL y mam parame d p ha: 
~ . . . . > ster oa ; or, Arc. ps l 
transformation can be used as the registration transformation ¢ ustenng ( : nena tana re Hyes 2 : dat 
> : SS. ; s eats - Perd Photogrammetry, Remote Sensing and Spatial Information 
function for scenes captured by satellite imaging systems with ue MT s f ma 
: - S i E Science, 34(3A):318-323. : 
narrow angular field of view. Moreover, 2-D similarity ity. 
i can oe Ud as T altemative for Some Tao, V. and Y. Hu, 2001. A Comprehensive Study for Rational 
ons wi ss anding requirements. : : : ; 
apo (cations à ess amende accuracy  TOQUTICTemts Function Model for Photogrammetric Processing, Journal of Th 
A similarity :neasure has been derived to ensure the > PPS an : 012): | 
: ; N Photogrammetric Engineering & Remote Sensing, 67(12): gra 
correspondence among conjugate line segments. The similarity 1347-1357 
measure does not assume that the end points of corresponding sax. doi 
eu d : : ad | J Ms : Th 
line. segments are identical. Finally, the primitives and the Wolfson, H., 1990. On curve matching. IEEE Transaction on . 
similarity measure are manipulated in a MIHT procedure to Pattern Recognition and Machine Intelligence, 12(5): 483-489. reli 
solve for the parameters involved in the registration pos 
transformation function while establishing the correspondence 9. ACKNOWLEDGMENTS 
between conjugate primitives. 
The author is grateful to Dr. Ayman Habib, Associate Professor ; 
Experimental results showed the feasibility and the robustness The University of Calgary, Department of Geomatics 
of the suggested approach that could tolerate possible Engineering for his scientific ^ suggestions and 
discrepancies between the imagery due to varying sensor recommendations. Author is also thankful to Mwafag Ghanma, [n3 
operational principles as well as changes in the object space. Graduate Research Associate, The University of Calgary. gh 
; . ; - > ; s : : 4 / 
Current research is focusing on automatic extraction of the Department of Geomatics Engineering for his kind help and in. 
registration primitives, affine transformation limitation and support. uh 
It 
9032