Understanding PTMs and RTIs
PTMs are generated by locking down a camera and an artifact and moving a light source around the two of them in know angles. For good results at least 30 different angles should be used. PATT (Photo Artifact TurnTable) uses 40 angles and images.
Let’s start by looking at light falling on the surface of a 3D artifact from a particular direction. Fig. 1 shows the artifact we will be using as an example1.
Fig. 1 Example Artifact
Fig. 2 Normal Reflectance on the surface of an artifact from a low angle light source
Fig. 2 shows light falling on the surface of an object from a low angle (Black Arrows). Light is scattered (Blue Arrows) in many directions. When viewed from above we see some of the scattered light and we see incised relief as the dark areas. As you can see with the lamp at this angle the small raised area on the right would be invisible as it is completely in a dark area. By moving the lamp around to many different angles and positions we create many different light and dark areas.
When you first view a PTM or RTI you will be viewing these light and dark areas with an added bonus. The program uses the 40 images that were taken with the lamp at 40 different positions and interpolates the information in such a way that, on screen, you can manipulate the lamp to virtually any angle.
This function alone can reveal much about an artifact, however, when Specular Enhancement is turned on even more detail might be seen. What happens is quite interesting. The light reflected undergoes a mathematical transformation (Polynomial Texture Mapping) which generates what is known as light normals. Since the artifact is three dimensional, as in Fig. 1, the reflected light scatters in many directions. The generation of a new, mathematically derived, light vector that is perpendicular to the surface of the 3D artifact and is of uniform intensity is called a normal. The program allows us to manipulate the normal vectors to reveal details in the surface texture of an artifact. This process gives the artifact a metallic-like appearance with very high contrast between the normals and the dark areas. Let’s look at Figures 3 and 4.
Fig. 3 Low Angle Light from the left transformed into normalized light vectors and its PTM
Fig. 4 Low Angle light from the right showing normalized light vectors and its PTM
Looking only at the Polynomial Texture Maps in Figures 3 and 4 we can see that a very high contrast images would result (Fig.5) and, via the software, and can be manipulated by changing which normals and dark areas that we are allowed to see. Once again, by interpolation, normals and dark areas of virtually any location can be viewed.
Fig. 51 The resultant images derived from the array of photos taken similar to PTM-1 and PTM-2 above
Figure 6, below, shows a summary of the image manipulations.
Polynomial Texture Maps and Reflectance Transformation Images represent complex mathematical processes2, however you do not have to be a mathematician to utilize them. PTM or RTI equipment like PATT and software can be tremendous additions to any archaeology research lab. The .lp file in Figure 6 is the Light Position Table that tells the PTM Fitter program exactly where the light was positioned in each of the 40 photographs.
Figure 7 is a closer view of the screen shot example from Figure 6.
Fig. 7 Screen Shot of the PTM Viewer image
References and for further reading
- Polynomial Texture Maps, Tom Malzbender, Dan Gelb, Hans Wolters
Hewlett-Packard Laboratories1 http://www.hpl.hp.com/ptm
- my incredible brain…lol