An Imaging Technique for Retrieval of Lost Content in Damaged Documents

Abstract

It is very common that some useful contents of documents are lost or hidden intentionally or accidently due to several reasons e.g. Whitener, pasting of paper, ink spreading, fading, dirt etc. The importance of these damaged documents may in terms of some research result, historic event, pacts or any important piece of information. The retrieval of such lost contents is a latent research area. The available approaches only guess for the lost contents by exploring the remaining intact information. But no approach is found yet to say that the retrieval is exactly of original ones. We have proposed a new approach and developed an experimental setup which involves imaging by sensing the light after passing through the damaged document instead reflection and then applying OCR on acquired images for retrieval of lost or hidden contents. Experiments are carried out on various test documents. Good results are obtained. The applicability of this scheme is limited for physically available documents only, but ensures the originality of retrieved contents.