anbul 2004 
ographical ER : 
grapes POLYGON-BASED TRUE ORTHOPHOTO GENERATION 
nentaritäts- 
RR Yevgeniy P. Kuzmin* , Sergey A. Korytnik", Orrin Long“ 
iversity o 
ja “ RL Software Technologies, Staten Island, NY, USA 
| digitalen " RL Labs, Moscow, Russia 
is of the - ZV Imaging Corporation, Huntsville, AL, USA 
-Wilhelms- 
f a digital KEY WORDS: Aerial images, Orthophoto images, elevation models, hidden areas detection 
data set. 
geospatial 
gart, 8-10 ABSTRACT: 
Obscured areas are serious defects in large-scale and urban orthophotos. In recent years several rectification schemes minimizing 
st Squares ; ; ; . : , 
these defects were proposed. These methods use raster-based algorithms for the detection of obscured areas and reconstruction of 
ARRIERE true orthophoto images. The common drawbacks of image-based approaches are that they require a lot of additional memory for 
intermediate structures and are slow and resolution dependent. In our paper we propose to use a polygon-based algorithm for the 
/, in: G..B. detection of obscured areas. Such a polygon-based approach is fast, accurate, and resolution independent. Spatial overlay processing 
> Decision provides the selection of the best orthophoto image for the recovery of detected obscured areas. The proposed method allows easy 
integration of true ortho image reconstruction techniques with existing commercial image processing software and spatial analysis 
Integration tools. 
:|len. Wiss. 
d DGK bei 
YD Thesis 1. INTRODUCTION 
1.1 Orthorectification Process 
GIS. ITC 
hede, The Original aerial images come in a perspective projection (a view 
from a camera focal point). For GIS usage, orthophoto images 
: : images in orthogonal projection) are needed. The process of 
integration tus e Pro) ) p 
transforming original images into orthoimages given an 
f the 21“ ; i - 
f accurate large-scale elevation model is illustrated at figure 1. 
1936-1942, 
edingungs- 
'errain and 
ystems. In: 
Computer 
inneapolis, 
      
      
          
1 
! 1 
! ' 
J i 
Orthoimage 
uthority of Y. Y zt 
nformation Fig 1: Illustration of orthorectification process 
de to LGN 
1 
1 1 
' 1 
i 
Y. y. 
M---- 
1.2 Double Mapping Effects 
  
In an urban landscape there are unavoidably hidden areas for 
which no information on the original aerial image exists. The 
presence of hidden areas results in double mapping effects on 
resulting traditional orthophoto images (see Figure 2 and Figure 
3). 
. n . . ^ 1 
Fig.2: Large scale orthorectified image with double mapping 
  
! All images are courtesy of BlomInfo A/S, Denmark 
$29