Historical documents dating using multispectral imaging and ordinal classification

Abstract

The estimation of the age of undated old manuscripts is one of the most challenging and controversial tasks in the field of historical document analysis. Several dating methods have been proposed, but most of them either use destructive techniques or rely on the textual information of documents. In this work, we rather focus our attention on the discoloration and the changes in the optical proprieties of their writing materials, which are a natural phenomenon that occurs as they age. Thus, we present a new content independent and non-destructive approach based on multispectral imaging combined with a ranking classification technique, to track the spectral responses of iron-gall ink at different wavelengths over time. We evaluated the proposed approach on multispectral images of real handwritten letters dating from the 17th to the 20th century. Experimental results demonstrate the effectiveness of multispectral imaging for document images dating.