Toposcopy produces either 3D textured or coloured 3D worlds or 2D CAD drawings of façades. At first the system was used only to accurately visualise a 3D design in a photo of the existing environment. As the system developed further it became possible to simultaneously measure and model existing buildings and other manmade objects using parametric models.

Economical Means
The means for carrying out a toposcopy survey are relatively cheap. The toposcope developed by us is independent of the availability of land-survey data and enables photos to be taken in a controlled manner. The basis of the system is an automatic level; an adapter connects the level with a digital camera. Together with the fieldwork program ‘Topo’ and a rod, the toposcope can be used as a simple tachymeter. All standard land-survey methods of determining a 3D point from angle and/or distance measurements are included. Photos are usually taken in a horizontal or upward direction When the top of a building to be modelled is not visible in the horizontal photo, then a second photo is taken in the same direction in the x/y plane, but this time with a positive vertical rotation angle. The accuracy of the data is sufficient for visualisations. However, for accurate photo-grammetric measurements we prefer to determine the position of camera and calibration points with a total station. In this case, the toposcope is still used to take photos in a controlled way and to do some supplementary angle measurements.

Visualisation Links
The main program is called ‘Scope’. Figure 1 shows the interface with loaded CAD drawing. Besides the general CAD toolbar at the top of the screen, there is also a bar with specialised tool buttons that automate procedures for making ground planes, extrusions and parametric models. The right-hand side of Figure 1 shows the calibration window, where the photo is being linked to the map. Clicking on points in the photo or on the map enters most data. Calibration varies according to the available hardware and data. When the camera point is not known, it can be calculated by clicking on three points on the map and in the photo. In such a case the angle of the viewing cone has to be estimated. When the camera point is located using a standard land-survey method, the viewing cone can be calculated instead. Orientation is determined using a 3D point known on the map and to be seen in the photo.

Photogrammetric Tools
When photos and map are linked together it is very easy to measure locations and heights by alternately clicking on points on the map and in the photo. The principles are simple and most routines work in both horizontal and upward photos.

• When the location of a point is known on the map it’s height can be determined with a single click in the photo.
  
• When a point lies in the vertical plane running through a line known on the map, it’s location and height can be determined with a single click in the photo.
  
• We can also measure like this in vertical planes that are at a certain distance from the known line.
  
• Only when an object is not shown on the map and it’s height is not known do we need two overlapping photos to determine both location and height.