1:250,000 sheet are clearly indicated at the larger scale. If maps are to be compiled from satellite 
image data to the standards used for the conventional 1:50,000 scale topographic sheets, comparable 
types and categories of details must be identifiable. 
To quantify this problem, three sets of image data were examined and planimetric maps (at 1:100,000 
scale) constructed for comparison with the 1:250,000 and 1:50,000 scale topographic maps (Figure 2). 
The image data sets included: 1) an Earth Terrain Camera S-190B photograph; 2) a Landsat-3 RBV image; 
and 3) a Landsat-3 MSS enhanced image product produced from a CCT. On the high resolution ETC photo- 
graph, the urban areas were well-defined and it was possible to identify four and two lane roads and to 
map street patterns. Airfields, which are high contrast targets, were easily mapped. However, 
forested areas were.difficult to delineate and small streams virtually impossible to detect. The 
1:100,000 scale planimetric map compiled from the ETC photograph is reasonably complete (Figure 2a). 
  
  
| 
(A LE, bate 
       
  
P 
Ve vr 
<q à 
  
  
  
  
  
  
  
c °°» > 
P : Lua X 
> 
5 10 y 
KILOME TRES KILOME TRES 
A B 
  
p ae 0m 
4-Lone Divided Highway A Urbon or Built-up Area 
——— 2-Lone Highway 
oo RalrOCÓ @ Forested Arec 
Power Tronsmission Line Cm Hydrologic Feature 
Lee 
  
  
  
Figure 2. Maps of the Trenton-Belleville, 
E Ontario area prepared from: : 
a) S-190B ETC photograph; b) Landsat-3 
RBV image; and c) enhanced Landsat-3 
MSS image. 
  
  
  
  
5 10 
KILOME TRES 
C 
On the RBV image, high contrast features such as built-up urban areas were easily delineated, as 
vere 4-lane highways. Some two lane roads were visible. Due to their low contrast, railroads and 
vegetated areas were difficult to detect. It was virtually impossible to locate small streams. The map 
compiled from the RBV image is incomplete (Figure 2b), particularly as compared to the map produced 
from the ETC photograph. 
A digitally enhanced false color composite MSS image depicted the urban areas as green patches. 
Forest lands, which appear bright red on the multispectral false color composite, were easily delineated. 
The spectral component helps to offset the poor spatial resolution. Major roads and some of the 
Secondary transportation features could be detected, but in general these features were poorly defined. 
The map compiled from the MSS image indicates that vegetation and stream patterns are more readily 
transferred from the multispectral data set (Figure 2c). 
In an effort to provide a quantitative comparison between the images and the information 
represented on the various maps, the categories of features were digitized from each product, including 
the 1:250,000 and 1:50,000 scale topographic maps. The time required to digitize a map is a surrogate 
51