Heritage3D

Laser scanning tests were carried out on an 18th-century gilded wood table (William Kent) at Chiswick House, London. Earlier tests had been carried out on the same table seven years earlier in 1999. The main aim of this work was to investigate the feasibility of laser scanning ornate gilded surfaces. The earlier work had highlighted the limitations of laser scanning such ‘shiny’ and complex surfaces. Results obtained in the 2006 survey were compared to those obtained in 1999 to evaluate the advances made in laser scanning during that period.

Instruments and software
Laser scanning was carried out using a 3D Scanners Modelmaker X laser scanning system. This comprises a scanning head mounted on an articulated arm, which is fixed to a heavyweight tripod. The scanning head is handheld, allowing accurate scanning of complex shapes. Point accuracy is specified as ±0.1 mm and working distance was approximately 10 cm. The raw scan data was processed using Innovmetrics Polyworks v. 9.1.7 software. Final post-processing was undertaken in Inus Technology Rapidform 2006 software.

Why was scanning selected?
The aim of this work was to investigate whether laser scanning can be used as a means of documenting in 3D ornate gilded surfaces and to compare the results with tests carried out seven years previously. Recording of such complex carvings using traditional techniques, such as hand-drawing, would be very difficult.

What problems were encountered?
Only minor problems were encountered, which were/would be straightforward to overcome: (1) Access to the tables was via a narrow spiral staircase; a portable tripod was therefore used instead of a heavy duty tripod. Page 2 of 2 This was not glued down for fear of staining the stone tiled floor, which meant extra care had to be taken not to over-extend the articulated arm, which could have resulted in slight movement of the tripod and hence the overall co-ordinate frame during scanning. (2) The shiny nature of the gilded surface meant that scanning was slightly slower than normal as more ‘passes’ of the scanner were required to capture data from the surface (part of the signal was reflected specularly away from the sensor rather than being scattered). (3) The marble table top prevented scanning some of the upper surfaces nearby. This could be overcome by removing the table top during scanning.

What were the final deliverables?
The raw scan data, processed computer models, metadata and a report. The results of this research clearly show that the extent and quality of data captured by laser scanning from the Chiswick tables has improved significantly since 1999. The research has also demonstrated that laser scanning can now be used to collect high quality 3D digital data from ornate gilded surfaces.