International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
Currency 
Old Recent 
Resolution 
  
Figure 6: Combination of quality attributes resolution (3D) 
and currency (color). 
4.2.2 Combination of Three Attributes: Color saturation 
can be used as a third graphical component to display 
another quality attribute of the data. The more saturated the 
color is, the better the data quality that it represents is. 
Figure 7 shows an example of how saturation can be added 
to express an additional attribute. Building on the example 
of figure 6 we can create the multi-variable color legend of 
figure 7. Variations from blue to green show timestamp 
information (x-axis) and changes in saturation depict 
available positional accuracies (y-axis). 
Recent Old 
High Time 
Low 
  
Pos. Accuracy 
Figure 7: Interaction of color hue and saturation. 
Figure 8 provides an example of using the legend of figure 7 
to add positional accuracy information to the graph of figure 
6. Note how the appearance of the graph changed, compared 
to figure 6. Figure 8 uses the same dataset for resolution and 
currency as figure 6, but also takes the positional accuracy 
of the data into account. 
An application of this visual combination method is the 
selection of appropriate information for cadastral updates. 
For this task, a town official needs the most recent 
information with high resolution and positional accuracy. 
All this information can be communicated using the single 
graph of figure 8. 
30. : > S 
FRE 
Resolution 
   
  
       
  
” 3800 
   
Figure 8: Combination of 3D, color, and saturation. 
4.2.3 Addition of Other Attributes: The previously 
discussed images offer an illustration of our visualization 
approach. In the included examples we convey resolution, 
currency, and positional accuracy. The remaining data 
quality attributes, namely completeness, scale and logical 
consistency, are conveyed using the following methods: 
* [ncompleteness of the data is shown by holes in the 3D 
illustrations, as can be seen in figure 9. To avoid the 
blocking of clear views of holes in the 3D surface, figure 
9 shows a top view. 
5500 
3600 
Figure 9: Top view of the visualization in figure 6 — white 
areas show incomplete data. 
* Scale can be depicted on the z-axis in lieu of resolution. 
* For communicating logical inconsistency in the data 
simple animations such as blinking can draw the user’s 
attention to the inconsistent area. 
5. CONCLUSIONS AND FUTURE WORK 
This paper presented a visualization environment for spatial 
image data quality, that provides an additional tool for users 
to browse and choose the right data for their tasks and 
improve their work efficiency. 
The data attributes positional accuracy, scale, resolution, 
completeness, consistency, and currency are visualized 
using a combination of 3D and color representations. We 
demonstrated how multiple attributes can be conveyed in à 
1006 
Internat 
Inr 
single ' 
color p 
availab| 
Part of 
advance 
user in 
data qu 
discuss 
feasibi 
explore 
This we 
through 
Intellig 
2008 a 
appreci 
Bastin, 
Uncert: 
Compu 
Beard, 
Craglia 
Resear 
pp. 28( 
Beard, 
Evaluaf 
et. al. ( 
and Te 
219-23 
Beard, 
NCGI/ 
Spatial 
Barbar: 
Beard, 
Quality 
30(2+3 
Butten 
Geogr: 
Hearns 
Geogr. 
Chiche 
Butten 
Cartog 
Butten: 
Qualit 
Baltim 
Claphz 
Visual 
Jose, C 
Davis, 
Multip 
23(4),