The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008 
To create a project by using ISPM 
- clutters (buildings, vegetation, water and bridges) 
- altimetry (ground height) 
- roads and railways 
Figure 1. Three Categories of Digital Data 
The client’s general requirements: To describe the clutter 
objects with a closed polygon; the linear objects with a 
continuous line. Each clutter object should be completed with a 
3D point located inside the polygon with a Z-value 
corresponding to the highest point of the object, and their height 
precision should be less than 1 meter. 
According the above-mentioned requirements, we complete the 
work in two steps: data capture and data editing. Here first we 
would like to introduce the method for data capture. 
4.1. Capture of Road Data 
4.1.1. The classification of road data 
It includes highways, main roads, secondary roads, local and 
urban routes, lanes and paths, railways. 
4.1.2. Geometrical specifications 
Because all the lines representing the roads networks contribute 
to the definition of digital terrain model, the density of the 
network can improve DEM. Therefore the 3D data component 
must be very accurate. The roads under and on the viaduct or 
river bridges must be captured in a continuous way; however 
they have to be interrupted in front of a tunnel entrance. The 
roads must intersect at common nodes, two intersecting roads 
must have the same z-value at intersection. 
4.2. Capture of altimetry data 
Altimetry is generally defined by the breaklines, but it can 
sometimes include 3D points and contour lines. The main 
purpose is to improve the digital terrain model outside the road 
network by giving all the terrain shapes and breaks. They 
include the tops of the embankments, the bottoms of the 
embankments, the thalweg lines, the crest lines, 3D mass points, 
contour lines. All these data should be captured with the same 
z-value at intersections. 
4.3. Capture of clutter data 
Clutter data are classified into four main categories: 
constructions, vegetation, water and bridges. At the end of 
capture, every kind of object must be described by a set of 
closed polygons whose height is given by a 3D point. This is 
performed in two main steps: 
Step 1: 3D capture 
3D capture, that means the exterior outlines of each object 
should be captured first and classified into different layer 
according to their height difference. A 3D point located within 
each clutter object should be captured at the maximal elevation 
of the object. It should be noticed that all the holes (yard, glade, 
or island) must be attached to the exterior contour using an 
interior limit. (See Figure 2) 
Step 2: Creation of topology (Generation of closed polygon) 
To transfer a set if 2D data from captured 3D data and all 
polygon objects will be picked up from 2D data to generate 
polygons by using GeoGraphics so as to reach an effect that 
each polygon must be associated to each 3D point. Please refer 
to Figure 3. 
(Figure 2) 
88