Exploring Character Shapes for Unsupervised Reconstruction of Strip-Shredded Text Documents

Paixão, Thiago M.; Bóeres, Maria Claudia Silva; Freitas, Cinthia O. A.; Oliveira-Santos, Thiago

doi:10.1109/tifs.2018.2885253

Cited by 12 publications

(30 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparative experiments showed that the accuracy of the proposed method (even in the modified version) was superior to the current state-of-the-art. When compared to Paixão [9], for instance, our method generalized better for documents with a more diverse layout and appearance, and also scaled more time-efficiently for the multi-document scenario. Furthermore, the time savings obtained by Marques [12] (based on the naive dissimilarity between edge pixels) were shown at the price of low reconstruction accuracy.…”

Section: Resultsmentioning

confidence: 99%

“…At a higher level of abstraction, compatibility can be assessed by exploring the matching degree of fragmented characters around the cutting section [9,10,19,31,32]. The continuity of character strokes was used as a matching criterion in [19,31].…”

Section: Related Workmentioning

confidence: 99%

“…The system described in Figure 1 (bottom pipeline) assumes that the documents were previously fragmented by a paper shredder and that the resulting shreds were scanned and separated in image files at a disk (the semi-automatic segmentation process adopted by our group is detailed in [9]). After loading data, the shreds are also binarized with Sauvola's algorithm [39] since the model was trained with binary samples.…”

Section: Reconstruction Of Mixed Shredded Documentsmentioning

confidence: 99%

“…This dataset was produced in the previous work of our research group [9,15] from a set of 20 business letters and legal reports of the ISRI-Tk OCR collection, the same set used in [25] to assess the reconstruction of simulated-shredded documents. The digital documents were printed onto A4 paper and subsequently submitted to a Leadership 7348 strip-cut paper shredder.…”

Section: S-isri-ocrmentioning

confidence: 99%

“…In this direction, a recent work [9] verified the lack of robust pairwise compatibility measures addressing real mechanically-shredded documents (i.e., cut by paper shredder machines) with low color information and predominant textual content. To fill this gap, the authors explored PR by using similarity between characters' shape as criteria for compatibility determination rather than traditional edge pixel-wise correlation, as used in [12][13][14].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Self-supervised deep reconstruction of mixed strip-shredded text documents

Paixão

Berriel

Bóeres

et al. 2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

The reconstruction of shredded documents consists of coherently arranging fragments of paper (shreds) to recover the original document(s). A great challenge in computational reconstruction is to properly evaluate the compatibility between the shreds. While traditional pixel-based approaches are not robust to real shredding, more sophisticated solutions compromise significantly time performance. The solution presented in this work extends our previous deep learning method for single-page reconstruction to a more realistic/complex scenario: the reconstruction of several mixed shredded documents at once. In our approach, the compatibility evaluation is modeled as a two-class (valid or invalid) pattern recognition problem. The model is trained in a self-supervised manner on samples extracted from simulated-shredded documents, which obviates manual annotation. Experimental results on three datasets-including a new collection of 100 strip-shredded documents produced for this work-have shown that the proposed method outperforms the competing ones on complex scenarios, achieving accuracy superior to 90%.

show abstract

Section: Resultsmentioning

confidence: 99%