See-through correction in recto-verso documents via a regularized nonlinear model

In this paper, we approach the removal of back-to-front interferences from scans of double-sided documents as a blind source separation problem. We consider the front and back ideal images as two individual patterns, overlapped in the observed recto and verso scans through a nonlinear convolutional mixing model. We adopt a regularization approach to estimate both the ideal images and the model parameters, by minimizing a suitable energy function of all the unknowns. The regularity of the solution images is described by typical local autocorrelation constraints, accounting also for well-behaved edges. This a priori information is particularly suitable for the kind of objects depicted in the images treated, i.e. homogeneous texts in homogeneous background, and, as such, is capable to stabilize the ill-posed, inverse problem considered. We show that the results obtained by this approach are much better than the ones obtained through data decorrelation or independent component analysis. As compared to approaches based on segmentation/classification, which often aim at cleaning a foreground text by removing all the textured background, one of the advantages of our method is that cleaning does not alter genuine features of the document, such as color or other structures it may contain. This is particularly interesting when the document has a historical importance, since its readability can be improved while maintaining the original appearance.