International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004 
  
  
   
  
   
   
   
   
   
  
  
  
  
   
   
    
  
  
   
  
   
  
  
  
  
  
  
  
  
  
  
   
  
   
   
  
  
  
   
  
  
  
  
   
   
  
  
  
  
  
   
  
   
   
   
   
   
   
   
   
   
   
  
  
    
2. THE MAP OF MEXICO 
The Map of Mexico 1550, also known as the Uppsala Map 
because it is kept at the Uppsala University Library, is one of 
the greatest existing historical documents recording the early 
colonial history of Mexico. The map is thought to be the work 
of the noted Spanish cosmographer Alonso de Santa Cruz. It is 
also one of only two known maps that give a fairly accurate 
picture of the ancient city of Mexico-Tenochtitlan and its 
surrounding regions in the mid 16" century. 
The clearly drawn roads over the mountains to other parts of the 
country permit us to retrace the routes taken by the Spanish 
conquerors. The map also gives information about the 
ethnography and the flora and fauna of the region. The 
population is shown performing a variety of activities, such as 
woodcutting, canoeing, hunting, and fishing. The approximately 
150 glyphs on the map, representing human and animal heads, 
feet, hands, circles and stars, refer to name places. (Larsson, 
2002) 
  
Figure 1. The Map of Mexico 
The dimensions of the document are 114 cm lengthwise and 78 
cm widthwise. Though no conclusive analysis of the material 
properties of the artifact has been done, expert opinion indicates 
that the inscriptions may have been painted using pigments on a 
surface comprising two sheets parchment joined in the center. 
(León-Portilla, 1986) During the years the parchment has 
deformed and acquired a three dimensional topography. 
3. REPRODUCTION OF THE MAP 
The requirements for the reproduction were that the three 
dimensional surface of the map must be taken into account and 
the resolution of the texture should be so high that also the 
smallest details of the map are distinguished. 
3.1 Image acquisition 
The original map was photographed in the University of 
Uppsala, Sweden. Prof. Henrik Haggrén from Helsinki 
University of Technology was developing a method for 
photographing maps and Dr. Lily Diaz-Kommonen and 
photographer Antti Huittinen from the University of Art and 
Design Helsinki took the photographs. 
In order to get enough detailed images, the map had to be 
photographed in pieces. The scale of the photographs was 
designed to be 1:2. In this scale the smallest details of the map 
can be distinguished using 5 x magnification. The used camera 
was a Sinar repro camera with a 180 mm macro Sinacron 
objective. The image size of the camera is 10 cm x 12,5 cm, so 
the map was photographed in pieces of about 20 cm x 25 cm. 
The photographs were taken through a grid frame, with grid 
size of 20 cm x 25 cm. The grid was made of thin threads and 
formed four strips, each strip consisting of six images. Each 
grid cell was covered with an image size of 4" x 5", color 
diapositive (100 ASA Kodak Ektachrome EPP, f /22.5) and 9 
cm x 12 cm, black and white negative. The distance between 
the camera and the surface of the map was about 66 cm. The 
grid was lifted a few centimeters up from the map so that the 
whole surface of the map was visible in the photographs. The 
grid was used here both for guiding the photography and for 
controlling the geometry between adjacent images. 
Because the surface of the map was three dimensional, an 
elevation model was needed for producing an image mosaic. 
Therefore each strip was recorded with a stereo photography of 
55 94 overlap. In order to provide "normal" stereo viewing of 
the map as well, an additional stereo mate was recorded for 
each image covering a full cell area. The base length was 22 
mm and the base-to-distance ratio was 1/30. This. corresponds 
to a normal stereo viewing at a distance of two meters. One 
additional set of b/w-images was recorded with low oblique 
lighting in order to produce relief effects by shadowing. In total 
80 color images and 124 black and white images were taken. 
(Haggrén) 
  
  
  
  
  
FT: $ @ 
— 66 cm 1 
A. A. A. Ah. A. A. A 
Y à d v à À v v à € 
fH i — 7 cm 
MM ——— ma 
Figure 2. The map was photographed in pieces through a grid 
frame. 
The images were scanned with pixel size of 1200 dpi or 20 um. 
Assuming that people normally see points having a size of 0.1 
mm, the scanned images could be viewed with 5 X 
magnification. 
When we got the scanned images it came out that the original 
edges of the images were cut off. It is obvious that this was a 
mistake because the original corners of the images would be 
needed for defining the interior orientation of the images. 
Despite of that we decided to use the images for the 
reproduction. 
   
Internat 
  
Figure 
32 0 
Totally 
were o 
were ci 
be use 
The re 
size, t 
would 
3.2.1 
the im 
points 
the pri 
not vei 
others 
why w 
be (0, 
orienta 
param 
3.2.2 
to one 
all the 
Totally 
measu 
Also 5 
image: 
pixels.