Restoration of historical RGB manuscripts via correlated component analysis.

In this paper, we consider the problem of removing interferences from pairs of recto-verso RGB documents affected by the see-through effect, due to either transparency or seeping of ink from the reverse side of the page. We formulate the problem as one of separating two individual texts, overlapped in the maps of the colour channels through an unknown linear mixing operator, and exploit statistical techniques of blind source separation. We show that recently proposed Correlated Component Analysis techniques overcome the already satisfactory performance of Independent Component Analysis techniques and colour decorrelation, when the two texts are even sensibly correlated.