B3. Istanbul 2004 
  
odel 
vith Table 1, it can 
surement greatly 
  
16.17 18 15 ZU 
'on-terrain points 
thorectified stereo 
(DTM) from each 
ssed by Baltsavias 
etermination from 
as that achievable 
al images if the 
-images is similar 
ixel size is similar 
Y, Z measurement 
DTM is with the 
rom stereomate 10 
DTM is a-rigorous 
curacy from the 
tely dependent on 
ied stereo images. 
  
  
     
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
That it to say, the measurement accuracy in the measurable 
seamless stereo model can reach a level as high as that of the 
ordinary stereo model if the pixel size of the mosaic orthoimage 
is similar with the scanning pixel size of the original aerial 
images. Those collusions are very valuable. It means that the 
measurement accuracy only relies on the orientation elements of 
the photos and the geometric quality of the photos but has 
nothing to do with DTM which is used for generating 
orthoimage and stereomate. This will make the measurable 
seamless stereo. model become more practical in practical 
applications, especially in these applications that the high 
measurement accuracy are required, because the acquisition of 
completely error-free DTM is a hard job (Li, et al., 1996). 
The experimental results obtained from this study are consistent 
with the above accuracy analysis. 
7. Conclusions 
In this paper, the concept of a measurable seamless stereo 
model is proposed and the method for generating such a model 
and algorithm for accurate 3D measurement on such a model 
are presented. The experimental results illustrate that high 
accuracy of 3D measurement in such a measurable seamless 
stereo model is achieved. If compared with the traditional 
methodology, it can be found that the improvement in accuracy 
is great, i.e. by 20 times. 
Though the stereo orthophoto pair is not a new concept, yet the 
measurable seamless stereo model is a new idea because it 
provides the possibility of additional function, i.e. accurate 3D 
measurement, apart from the seamless stereo view and 
traditional 3D measurement from the seamless model based on 
an orthoimage and its stereomate. The high accurate 
measurement is achieved by making use of the stereo pairs, 
through recording the lineage of each pixel in the mosaic 
orthoimage and mosaic stereomate. 
From practical point of view, the photogrammetric system with 
such a measurable seamless stereo model function can be easily 
integrated into GIS. Such an integrated system provides a new 
practical and economical approach to stereoscopically render 
terrain features in a large area and allow users to carry out 
accurate three. dimensional measurement of any objects and 
features visible for 3D information acquisition, data updating 
and/or other applications such as 3D design, analysis and 
photo-interpretation, in many fields such as planning, forestry, 
electric power, geology and so on. 
References 
Baltsavias E.P. , 1996. Digital ortho-images--a powerful tool for 
the extraction of spatial- and geo-information. ISPRS 
Journal of Photogrammetry & Remote Sensing, 51(2), pp. 
63-77 
Blachut, T.J., 1968. Further Extension of the Orthophoto 
Technique. The Canadian Surveyor, 22(1), pp. 206-220. 
Blachut, T.J. and van Wijk, M.C., 1970. 3D information from 
orthophotos. Photogrammetry Engineering, 36(4), pp. 
365-374. 
Blachut, T.J., 1976. The Stereo-Orthophoto Technique in 
Cadastral and General Mapping. Photogrammetric 
Engineering and Remote Sensing, 42(12), pp.1511-1519. 
Collins, S.H., 1968. Stereoscopic orthophoto maps. Canadian 
Surveyor, 22(1), pp. 167-176. 
Collins, S.H., 1969. The Accuracy of optically projected 
orthophotos and stereo-orthophotos. The Canadian Surveyor, 
24(5), pp. 450-463. 
Collins, S.H., 1970. The ideal mechanical parallax for 
sterro-orthophoto. The Canadian Surveyor, 24(5), pp. 
561-568. 
Collins, S.H., 1972. The  stereo-orthophoto 
Photogrammetric Engineering, 38 (12), pp.1195-1202. 
Collins, S.H. and van Wijk, M.C., 1976. Production and 
Accuracy of Simultaneously Scanned Stereo-orthophoto. 
Photogrammetric Engineering, 42(12), pp.1521-1528. 
Kraus, K., Otepka, G., Lottsc, J. and Haitzmann, H., 1979. 
Digitally Controlled Production of Orthopotos and 
Stereo-Orthopotos. Photogrammetric Engineering and 
Remote Sensing, 45(10) , pp. 1353-1362. 
Kraus, K., 1984. Photogrammetrie. Band2, F. Duemmlers 
Verlag, Bonn. 
Li, D. R., Wang, S. G. and Li, R. X., 1996. Automatic quality 
diagnosis in DTM generation by digital image matching 
techniques. GEOMATICA, 50(1), pp. 65-73 
Li D. R., Wang M., Gong J. Y., 2002. Principle of seamless 
stereo orthoimage database and its measurement accuracy 
analysis. Proceedings of the ISPRS Commission III 
Symposium, Graz, Austria, pp: 151-156 
Sarjakoski, T., 1990. Digital stereo imagery--a map-product in 
the future? Proceedings of the ISPRS Commission IV 
Symposium, Tsukuba, Japan, pp.200-210. 
van. Wik, MC, 1979. Geometrical quality of 
stereo-orthophotos produced from automatic image 
correlation data. Photogrammetric Engineering and Remote 
Sensing, 45(10), pp.1363-1369. 
Wang M., 2001. The development of large-scale seamless image 
database (GeolmageDB) and the feasible research on 
measurable virtual reality. Ph.D. dissertation, LIESMARS, 
Wuhan University, China. 
Wang Z. Z., 1990. Principles of photogrammetry (with remote 
sensing). Publishing House of Surveying and Mapping, 
Beijing, China. 
pair. 
Acknowledgements 
The research described in this paper was funded by the Youth 
Foundation of National Nature Science Committee (No. 
40201044). This support is valuable. The author would like to 
thank Prof. Deren Li and Prof. Zhilin Li for their constructive 
comments.