Analysis on optical coherence tomography images to detect irregularities and restoration on paper-based artwork

Image analysis techniques were developed and used on optical coherence tomography (OCT) images for the analysis of restoration on paper-based artwork. Three etchings were imaged using an 840 nm portable OCT system from Lumedica and analysed using an algorithm developed to automatically detect restoration boundaries. The culmination of the algorithm was in its use for analysis on Landscape with an Obelisk, produced in the 1650s by Rembrandt van Rijn. The artwork is of interest due to the extensive restoration work conducted by art restorer Carl Schweidler in the 1920s, renowned for his restoration technique of carefully shaving the edges of both artwork and repair pieces before piecing them together perfectly. The data processing pipeline was optimised using three etchings that contain drastically different scales of restoration work so that the analysis algorithm may be used for future paper-based conservation studies.