Full text: Proceedings, XXth congress (Part 3)

z (DG), 
y actual 
Chinese 
es to the 
CASM) 
uctivity. 
ina was 
1:4,000 
on were 
ophotos 
inology. 
100,000 
netry in 
scussed. 
  
Yo 
Yo 
  
  
  
"China 
| in the 
China. 
y since 
n days. 
ay find 
ed. It is 
imunity 
Ithough 
ountry. 
ologies 
; in the 
of the 
ping is 
digital 
igitized 
duction 
:edures 
rs, and 
lready. 
ency of 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
  
production and intensive human work is still required, e.g. 
photographs have to be digitized, cartographic features should be 
extracted, and ground control points (GCP) must be collected and 
measured in the field. Image digitization can be accelerated 
considerably by adding some automatic mechanisms on the film 
scanner, or ignored completely when digital aerial cameras are 
applied. Feature extraction can profit a lot [rom DPWS automated 
functions being improved gradually with the current technical 
progress. Now, the task of GCP collection and measurement 
remains the inevitable labor work, which slows down the 
production directly on the one hand, and makes mapping of e.g. 
desert and inaccessible regions in west China very hard or even 
impossible on the other. 
Thus, aerial photogrammetry without any ground control or with 
a limited number of GCP is the prerequisite for speeding up the 
map production so as to meet the demands of rapid development 
in China and closing the map gap in the country's western regions, 
in particular. 
2. DGPS/IMU-BASED PHOTOGRAMMETRY 
Recent developments in Differential Global Positioning System 
(DGPS) and Inertial Measurement Unit (IMU) technologies make 
direct measurement of sensor Exterior Orientation parameters 
(EO) possible, so as that the image orientation can be without or 
just with a limited number of GCP determinable. In this kind of 
DGPS/IMU-based photogrammetry the traditionally dominating 
Aerial Triangulation (AT) becomes now a supporting role in the 
image orientation. Depending on how AT is used in the whole 
procedure, we can distinguish a Direct Georeferencing (DG) from 
an Integrated Sensor Orientation (ISO) conceptually. 
2.1 Direct Georeferencing 
Use of the image EO determined based on direct DGPS/IMU 
measurements for photogrammetric data processing without 
conducting an AT over the entire image block presents the 
principle of Direct Georeferencing. Advantages herewith are 
obvious: No tie point determination (either manual or automatic) 
is necessary, no GCP is required and, as a matter of fact, no AT 
has to be performed for the entire image block anyway. 
Nevertheless, for purpose of a so-called boresight alignment 
highly accurate AT results of a small calibration field are still 
needed to determine the attitude differences between the image 
sensor coordinate system and the IMU one (e.g. Cramer, 2002). 
Some simple geometrical considerations (Kremer, 2002) show 
that DG is quite suitable for small and medium scale mapping, 
especially for orthoprojection, and for large scale mapping, 
however, there may be certain limitations using this approach. 
2.2 Integrated Sensor Orientation 
Simultaneously processing DGPS/IMU data and image 
information in a combined AT over the entire image block can be 
referred to as Integrated Sensor Orientation. Obviously, this 
approach combines advantages of conventional AT and modern 
DG, and compensates for their weaknesses. For instance, based 
on tie points well-distributed over image areas and GCP over the 
block AT is capable of eliminating systematic errors by means of 
a self-calibrating adjustment, and thus reaches the highest 
accuracy and stability of image orientation. With the direct EO 
871 
measurements provided by the DGPS/IMU approach the best 
initial values are available for both the bundle block adjustment 
and the automatic tie point determination in an automatic AT (e.g. 
Tang, 1999). Fully profited from the stable geometry supported 
by the direct EO measurements the number of GCP required for 
AT can be reduced to the minimum, and the performance of the 
entire image orientation procedure is tremendously improved 
after all. 
What a kind of approach to be chosen for a specific project 
depends on many factors, e.g. the accuracy required, the image 
scale, the availability of ground control information and GPS base 
stations and accessibility of the project area. For large scale 
mapping projects it would be better to use the integrated sensor 
orientation, in general. 
3. FIRST PROJECT 
With the successful import of a CCNS/AEROcontrol system 
(Grimm, 2003) in November 2002, Chinese Academy of 
Surveying and Mapping (CASM) organized the first 
DGPS/IMU-based photogrammetric project in China in 
cooperation with the Germany-based companies IGI and 
Techedge. Anyang, a city of Province Henan, Located about 500 
km south of Beijing, was chosen for the test flights. CASM was 
responsible for all administrative as well as operational issues of 
the project. The German team, consisting of the authors, was in 
charge of workflow design, system installation, personal training 
and technical consulting within the project. 
3.1 Project Realization 
Two missions of aerial photography were planned and conducted 
for a photo scale of 1:4,000 (for 1:1,000 scale mapping) and 
1:20,000 (for 1:5,000 scale mapping), respectively. Concerning 
many special cases in China (e.g. inaccessible regions) 
investigating effects of GPS base station location was also one of 
the objectives. Thus, several GPS base stations of different 
distances to Anyang were simultaneously used for the missions. 
Table 2 shows the major mission parameters. 
  
  
  
  
  
  
Mission A Mission B 
Photo scale 1:4,000 1:20,000 
Map scale 1:1,000 1:5,000 
Terra Flat Flathilly 
type 
Area 5 km x 2.5 km 14 km x 22 km 
Block size | 5 strips x18 photos 7 strips x16 photos 
  
Calibration [Inside block, additional outside block, 2 strips of 11 
  
  
  
  
  
  
  
  
  
  
  
field 11 photos photos each 
GSP base ebur erit ; A Em 
m Anyang| Hebei Beijing Anyang|Zhengzhou| Beijing 
station 
Distance 10 | _ |212kmlésikm| - | 150km |451 km 
block 
Flight date | 18/11/02 | 19/11/02 03/01/03 
  
Table 2. Major mission parameters of the Anyang project. 
3.2 Mission A 
To prove the applicability of the DGPS/IMU technology for 
 
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.