Assessing cross-cut shredded document assembly

In this paper we address the problem of quantitative evaluation of<br>cross-cut shredded document reconstruction.<br>We propose quantitative metrics using graph theory and classic information<br>retrieval concepts to compare the neighborhood connectivity graph of a reassembled<br>document shredded by a cross-cut machine against the neighborhood graph of the ground-truth.<br>These metrics focus entirely on the proper relative positioning of the shredded pieces.<br>To do so, we have shredded 12 documents containing diverse content, such<br>as handwriting, printed text, images and photographs. We then scanned, extracted<br>the pieces, and reassembled them into the ground-truth.<br>This dataset is available to the readers, with the original documents,<br>the digital representation of the shreds, and the scripts that provide<br>the quantitative evaluation of the user’s reconstructions.