International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004 
  
^ 
cid e 
v RT M 
xt ea AY E 
Cati Ties 
a vo 
  
  
  
   
  
  
  
    
  
Fermé tire À he rar ii 
Figure 10. The Experimental Result Corresponding with 
Figure.9 
In Fig.10, we can see obviously the trend of the relation 
between data amount and information amount in the 
background of a whole map. When the data amount is little, the 
information amount is little too, which is obvious because a 
small bulk of data cannot obtain much information. When the 
data amount increases, the information amount increases in a 
common sense, but what we must pay attention to is that a 
relatively large amount of data does not necessarily contain 
more information. With a certain amount of data, we can get 
different amount of information, but the largest amount of 
information is related to the data amount. 
When the data amount increases continually, a turning point is 
achieved. After this turning point, when the data amount 
increases, the maximal information amount that contained by 
the corresponding data amount do not increase. This turning 
point suggests that to a certain resolution and a window with a 
given size, there exists an upper limit of information amount. 
To get this upper limit of the information amount, we need a 
certain amount of data, if more data is given, we cannot get 
more information. This conclusion is what we can get from the 
experimental research and it is accord with what we proposed in 
Section 3. 
5. CONCLUSIONS AND FUTURE WORK 
In this paper, the existing quantitative measures for map 
information are first briefly presented and their common 
shortage is pointed out. All the existing methods are 
probability-based and have not reveal the relation between the 
amount of data and information, which is the major problem 
that this paper solves. From the viewpoint of QoS in the 
distributed environment, a new raster-based method of map 
information measurement is proposed and its rule in the 
dissemination of spatial information in the Internet is described. 
Finally an experimental research result is given, and the relation 
between the so-called raster-based information amount and the 
data relevant to it is analysed exactly with some statistical 
method. 
What we present in this paper is an initial work about the 
application of map information measurement in spatial 
information service and there are more work left to do in the 
future. We can expect that to certain kinds of map, such as 
contour lines and road networks there must exist some 
difference among their data-information amount relationships 
and it is an interesting region for future research. The quantified 
comparison between this information measurement and other 
information measurement can be done more profoundly and 
370 
more valuable results are expected to be found. More work will 
be done about this new method of information measurement. 
References: 
SHANNON, C. E, 1948, A mathematical theory of 
communication. The Bell System Technical Journal, 27, 379. 
423 & 623-656. 
SUKHOV, V. L, 1970, Application of information theory in 
generalization of map contents. /nternational Yearbook of 
Cartography, X , 41-47. 
SUKHOV, V. L., 1967, Information capacity of a map entropy. 
Geodesy and Aerophotography, X, 212-215. 
NEUMANN, J., 1987, Gnoseological aspects of improving the 
geometrical component of the space-time model in cartography. 
In Proceedings, 13" International Cartographic Conference 
ICA, Morelia, Mexico, IV: pp.569-584. 
Bjorke J. T., 1996, Framework for entropy-based map 
evaluation. Cartography and Geographical Information 
Systems, 23, 78-95. 
Li Z. L. and HUANG P. ZH., 2002, Quantitative measures for 
spatial information of maps, INT. J. Geographical Information 
Science, Vol. 16, No. 7, 699-709. 
KE’ 
ABS 
This 
prov 
into 
info 
our | 
tour 
and 
com 
only 
Rec 
GIS 
Wet 
disse 
their 
acce 
distr 
map: 
to g 
that 
this 
para 
imag 
conn 
Clien 
  
Figui 
1997