Surface Displacement Measurements of Artworks: New Data Processing for Speckle Pattern Interferometry

Abstract

Curators have developed preventive conservation strategies and usually try to control the temperature (T) and relative humidity (RH) variations in the museum rooms to stabilise the artworks. The control systems chosen by museums depend on the size and age of the building, the financial means and the strategies that can be adapted. However, there is a lack of methods that can monitor mechanical changes or chemical reactions of objects in real-time or regularly. It would therefore ideally be preferable to monitor each of them to alert them to preserve them. For this purpose, a non-destructive, non-contact, full-field technique, Digital Holographic Speckle Pattern Interferometry (DHSPI), has already been developed and allows direct tracking of changes on the surface of artworks. This technique is based on phase-shifting speckle interferometry and gives the deformation of the surface below the level of the micro-meter of the analysed object. In order to monitor the deformation continuously, a large number of images are acquired by DHSPI and have to be processed. The existing process consists of removing noise from the interferogram, unwrapping this image, and deriving and displaying a 2D or 3D deformation map. In order to improve the time and accuracy of processing the imaging data, a simpler and faster processing method is developed. Using Matlab®, a denoising methodology for the interference pattern generated during data acquisition is created, based on a stationary wavelet transform. The unwrapped image is calculated using the CPULSI (Calibrated Phase Unwrapping based on Least-Squares and Iterations) algorithm as it gives the fastest results among the tested methods. The unwrapped phase is then transformed into surface displacement. This process performs these steps for each interferogram automatically. It allows access to 2D or 3D deformation maps.