Learning to Reassemble Shredded Documents

Richter, Fabian; Ries, Christian X.; Cebron, Nicolas; Lienhart, Rainer

doi:10.1109/tmm.2012.2235415

Cited by 32 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, vertical dissimilarity between two pieces x i and x j , where x i is placed on the top of x j is defined as (2) where square difference between pixels of the last row of x i and the first row of x j is computed. These measures of dissimilarity will be calculated many times during the process of evaluation of thousands of individuals and therefore it is advantageous to create a lookup table covering all the relevant values at the beginning of run.…”

Section: Problem Definition and Representationmentioning

confidence: 99%

“…It is a well-known fact that jigsaw puzzles and their equivalents belong to NP-complete class [1] and thus provide serious obstacles for various attempts to solve them automatically. Moreover, there are many practical applications ranging from image mosaicking up to the reconstruction of archaeological artifacts and assembly of shredded documents [2]. The first jigsaw puzzle was produced in 1760 in London [3] and then the game of puzzle gradually expanded all over the world.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sequence Matching Genetic Algorithm for Square Jigsaw Puzzles

Hynek

2014

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

View full text Add to dashboard Cite

Part 7: Genetic AlgorithmsInternational audienceOur paper presents a new method for solving the rectangle piece jigsaw puzzle problem. The puzzle image is RGB full color and because of uniform shape of the individual pieces the process of puzzle assembly is based on information of the pixel values along the border line of the piece only. We have utilized a genetic algorithm that searches for the optimal piece arrangement using dissimilarity between adjacent pieces as the measure of progress. Unlike the previous attempts to utilize genetic algorithms to solve the problem, we have proposed a new heuristic asexual operator that aims at identification of points of fraction within partially assembled picture, extraction of supposed sequence of correctly joint pieces, and its insertion into a new position in such a way that, if possible, the segment is enlarged. Our approach has been successfully tested and the algorithm is capable of solving puzzles consisting of several hundred pieces

show abstract

Section: Problem Definition and Representationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Sequence Matching Genetic Algorithm for Square Jigsaw Puzzles

Hynek

2014

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

View full text Add to dashboard Cite

show abstract

“…Other recent works investigate on the automatic assembly of documents [7] and photos [8]. The latter approach of Cao et al deals with the problem of simultaneously reconstructing multiple photos from a collection of pieces.…”

Section: Related Workmentioning

confidence: 99%

“…In this work we use the bdw082010 dataset [7], which consists of 96 document pages that have been shredded by hand into 16 pieces each. Our preprocessing closely follows [7] to obtain an approximation of each fragment's contour.…”

Section: Dataset Of Shredded Documentsmentioning

confidence: 99%

Partial contour matching for document pieces with content-based prior

Richter

Ries

Romberg

et al. 2014

2014 IEEE International Conference on Multimedia and Expo (ICME)

Self Cite

View full text Add to dashboard Cite

In this paper we present a method for aligning shredded document pieces based on outer contours and content-based prior information. Our approach relies on domain-specific knowledge that document pieces must complement each other when aligned correctly. Building on this intuition we propose a variant of MSAC (M-estimator SAmple Consensus) to estimate an hypothesis that recovers the spatial relationship between pairs of pieces. To do so we first approximate their boundaries by polygons from which we define consensus sets between fragments. Each consensus set provides multiple hypotheses for aligning one piece onto the other. An optimal hypothesis is identified by applying a two-stage procedure in which we discard locally inconsistent hypotheses before verifying the remainder for global consistency.

show abstract

“…In the last decade, many studies focused on the reconstruction of documents shredded by hands [6][7][8][9], and by strip-cut machines [10][11][12]. In the case of cross-cut shredded document reconstruction, existing approaches have initially been tested only on simulated shredded text documents [13][14][15][16][17], in which the shreds have a rectangular shape and its corners are aligned perfectly.…”

Section: Related Workmentioning

confidence: 99%

Assessing Cross-Cut Shredded Document Assembly

Saboia

Goldenstein

2014

Advanced Information Systems Engineering

View full text Add to dashboard Cite

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

show abstract

Learning to Reassemble Shredded Documents

Cited by 32 publications

References 19 publications

Sequence Matching Genetic Algorithm for Square Jigsaw Puzzles

Sequence Matching Genetic Algorithm for Square Jigsaw Puzzles

Partial contour matching for document pieces with content-based prior

Assessing Cross-Cut Shredded Document Assembly

Contact Info

Product

Resources

About