Reassembling Shredded Document Stripes Using Word-Path Metric and Greedy Composition Optimal Matching Solver

Liang, Yongqing; Li, Xin

doi:10.1109/tmm.2019.2941777

Cited by 11 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…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. Finally, the recently published Liang method [33] performed significantly worse than the proposed method in terms of accuracy, in addition to a limited time-scalability to realworld scenarios comprising several documents.…”

Section: Discussionmentioning

confidence: 89%

“…Therefore, reconstruction with this method was limited to k = 5 documents. The second method is the one proposed by Liang and Li [33], which is referred to as Liang. Due to time restrictions of the provided implementation 4 , the multi-reconstruction experiment was run only for the datasets S-Marques and S-Isri-OCR limited to k = 3 documents.…”

Section: Methodsmentioning

confidence: 99%

“…In recent work, Liang and Li [33] proposed the word path metric, which combines pixel-and character-level information (low-level metrics) with word-level information (high-level metric). A central procedure in their method is sampling candidate sequences and applying OCR for word recognition to improve pair compatibility estimation.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

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

Paixão

Berriel

Bóeres

et al. 2020

Pattern Recognition

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: Discussionmentioning

confidence: 89%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

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

Paixão

Berriel

Bóeres

et al. 2020

Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…It has been proposed as well for other miscellaneous tasks such as invoice analysis [38], pairing different versions of historical manuscripts [19], and reassembling shredded document stripes [27], to name a few. All these works are completely unrelated with evaluation of HTR transcripts, which is the topic of this paper.…”

Section: Metrics Related With the Hungarian Algorithmmentioning

confidence: 99%

End-to-End Page-Level Assessment of Handwritten Text Recognition

Vidal¹,

Toselli²,

Ríos-Vila³

et al. 2023

Preprint

View full text Add to dashboard Cite

The evaluation of Handwritten Text Recognition (HTR) systems has traditionally used metrics based on the edit distance between HTR and ground truth (GT) transcripts, at both the character and word levels. This is very adequate when the experimental protocol assumes that both GT and HTR text lines are the same, which allows edit distances to be independently computed to each given line. Driven by recent advances in pattern recognition, HTR systems increasingly face the end-to-end page-level transcription of a document, where the precision of locating the different text lines and their corresponding reading order (RO) play a key role. In such a case, the standard metrics do not take into account the inconsistencies that might appear. In this paper, the problem of evaluating HTR systems at the page level is introduced in detail. We analyze the convenience of using a two-fold evaluation, where the transcription accuracy and the RO goodness are considered separately. Different alternatives are proposed, analyzed and empirically compared both through partially simulated and through real, full end-to-end experiments. Results support the validity of the proposed two-fold evaluation approach. An important conclusion is that such an evaluation can be adequately achieved by just two simple and well-known metrics: the Word Error Rate, that takes transcription sequentiality into account, and the here re-formulated Bag of Words Word Error Rate, that ignores order. While the latter directly and very accurately assess intrinsic word recognition errors, the difference between both metrics gracefully correlates with the Spearman's Foot Rule Distance, a metric which explicitly measures RO errors associated with layout analysis flaws.

show abstract

“…A database containing 90000 kinds of words is generated by artificial synthesis. Each word is regarded as a class, and 90000 words are recognized by CNN [6]. The obvious disadvantage of this algorithm is that it can not recognize words outside the dictionary, and due to the fixed size of the input image, extrusion deformation will occur when the length of the word is too long, which will affect the accuracy of recognition.…”

Section: Recognition Algorithm Based On Character/wordmentioning

confidence: 99%

Nameplate text recognition model based on attention mechanism

Wu,

Wang,

Yang

2023

Third International Conference on Artificial Intelligence, Virtual Reality, and Visualization (AIVRV 2023)

View full text Add to dashboard Cite

With the accelerating speed of power grid construction, the number of substations is increasing, and the management of power equipment is becoming more and more complex. The traditional manual management method is time-consuming, labor-consuming and inefficient. Due to the complex scene of power equipment nameplate, the failure of standardization of nameplate size, many characters, miscellaneous types and many steel seal information in the nameplate, the recognition accuracy of power nameplate needs to be improved. On the basis of fully mining the principle of attention mechanism encoder and decoder, combined with LSTM and convolutional neural network, this paper proposes a nameplate text recognition model based on attention mechanism to solve the problem of automatic nameplate text recognition. Experiments show that the method proposed in this paper can effectively identify the substation nameplate.

show abstract

Reassembling Shredded Document Stripes Using Word-Path Metric and Greedy Composition Optimal Matching Solver

Cited by 11 publications

References 38 publications

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

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

End-to-End Page-Level Assessment of Handwritten Text Recognition

Nameplate text recognition model based on attention mechanism

Contact Info

Product

Resources

About