Laser Surface Scanning
Laser surface scanning is a deceptively simple process. The scanner generates a laser line that is directed towards that object. Various parts of this vertical line hit different parts of the object at different distances and the light is reflected back to the laser scanner at a collector that keep track of the time it takes for the return trip – the faster the time the close that part of the object is. It takes a very precision device to accomplish this and the scanners we have can give a resolution of up to 0.1 mm. The specimen is rotated in front of the laser source to give readings across the its whole three-dimensional shape.
Building virtual models is a relatively long process, depending on the complexity of the shape of the object being scanned. Simple shapes, like plaster models of footprints, take relatively little time (e.g., an hour or less) but complex skulls or sacra can take days to produce an accurate model. The specimen is placed in front of the scanner in a manner that will maximize the success of the scanner – the first scan is often a vertical position. The scanner then takes readings at a selected angle interval, the more readings the higher the accuracy of the model. These are then merged by the software into a single mega-scan. These scans inevitably will miss some information for regions tangential to the laser source, so additional scans are done using other positions and these mega-scans are then merged by our scanning technicians to get all the surface morphology.
Data editing is essential as there are usually lots of spurious points generated during the editing process. Luckily, these tend to be concentrated in certain regions, especially areas not easily accessed by the laser line from that angle. These individual points are edited out by the scanning technician. This can include whole regions that will be better scanned from a different angle. With the software available to us, the merging of the mega-scans can be easy or it can take a lot of effort by the scanning technician. This inevitably involved rotating one set of data until it properly fits another and takes a lot of experience to do well. Happily, the process results in spectacular models.
The scanning process takes a lot of time to learn and there are ways to make it run more smoothly. First, you must pay attention to the color and reflectance of the specimen as it will greatly affect scan quality. Black glass – such as the black obsidian present in many of the projectile points we scan – is invisible to the laser and will just leave blank space. We solve this problem typically with a light dusting of neutral talcum powder, which can give spectacular results. For vertical scans, it is best to not do it straight up-and-down but with a small angle. Further subsequent scans should not be done at ninety degrees to the first but with more overlap to assist the merging process of the mega-scans.