Progressive Alignment and Discriminative Error Correction for Multiple OCR Engines

Lund, William B.; Walker, Daniel D.; Ringger, Eric K.

doi:10.1109/icdar.2011.303

Cited by 20 publications

(23 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Klein et al (2002) show that combining complementary result from different OCR models leads to a better output. Lund et al (2011) demonstrate that the overall error rate decreases with the addition of different OCR models, regardless of the performance of each added model. Lund et al (2013a) use machine learning techniques to select the best word recognitions among different OCR outputs.…”

Section: Related Workmentioning

confidence: 86%

“…One direction of work ensembles outputs from multiple OCR engines for the same input and selects the best word recognition as the final output (Klein et al, 2002;Cecotti and Belayd, 2005;Lund and Ringger, 2009;Lund et al, 2011;Lund et al, 2013a;Lund et al, 2013b). Klein et al (2002) show that combining complementary result from different OCR models leads to a better output.…”

Section: Related Workmentioning

confidence: 99%

“…While machine learning approach exhibits advantages in correcting OCR-generated texts, two problems emerge from the existing models: First, some models (Lund and Ringger, 2009;Lund et al, 2011;Lund et al, 2013a;Lund et al, 2013b;Kissos and Dershowitz, 2016) limit candidate suggestions from the recognition output of OCR engines. The errors unrecognized by all OCR engines are thus unable to be corrected.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Statistical learning for OCR error correction

Mei

Islam

Moh’d

et al. 2018

Information Processing & Management

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 86%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Statistical learning for OCR error correction

Mei

Islam

Moh’d

et al. 2018

Information Processing & Management

View full text Add to dashboard Cite

“…In conclusion this paper has explored the use of in-domain training; where in-domain data is not available, techniques for using out-of-domain training data such as those explored in Lund, Walker, and Ringger (2011) [9] can be used.…”

Section: Discussionmentioning

confidence: 98%

Error Correction with In-domain Training across Multiple OCR System Outputs

Lund

Ringger

2011

2011 International Conference on Document Analysis and Recognition

Self Cite

View full text Add to dashboard Cite

Optical character recognition (OCR) systems differ in the types of errors they make, particularly in recognizing characters from degraded or poor quality documents. The problem is how to correct these OCR errors, which is the first step toward more effective use of the documents in digital libraries. This paper demonstrates the degree to which the word error rate (WER) can be reduced using a decision list on a combination of textual features across the aligned output of multiple OCR engines where in-domain training data is available. This research was performed on a data set for which the mean WER across the three OCR engines employed is 33.5%, and the lattice word error rate is 13.0%. Our correction method leads to a 52.2% relative decrease in the mean WER and a 19.5% relative improvement over the best single OCR engine, as well as an improvement over our previous work. Further, our method yields instances where the document WER approaches and for five documents matches the lattice word error rate, which is a theoretical lower bound given the evidence found in the OCR.

show abstract

“…In order to further validate the utility of the synthetic datasets, we used the data to train an existing approach to multipleengine OCR error correction, as described by Lund et al 14 The method is motivated by the reasoning that OCR engines have different strengths and weaknesses. Thus, if one OCR engine outputs an incorrect hypothesis for a word token in the source image, another engine might output the correct hypothesis.…”

Section: Ocr Error Correction Taskmentioning

confidence: 99%

A synthetic document image dataset for developing and evaluating historical document processing methods

2012

View full text Add to dashboard Cite

Document images accompanied by OCR output text and ground truth transcriptions are useful for developing and evaluating document recognition and processing methods, especially for historical document images. Additionally, research into improving the performance of such methods often requires further annotation of training and test data (e.g., topical document labels). However, transcribing and labeling historical documents is expensive. As a result, existing real-world document image datasets with such accompanying resources are rare and often relatively small. We introduce synthetic document image datasets of varying levels of noise that have been created from standard (English) text corpora using an existing document degradation model applied in a novel way. Included in the datasets is the OCR output from real OCR engines including the commercial ABBYY FineReader and the open-source Tesseract engines. These synthetic datasets are designed to exhibit some of the characteristics of an example real-world document image dataset, the Eisenhower Communiqués. The new datasets also benefit from additional metadata that exist due to the nature of their collection and prior labeling efforts. We demonstrate the usefulness of the synthetic datasets by training an existing multi-engine OCR correction method on the synthetic data and then applying the model to reduce word error rates on the historical document dataset. The synthetic datasets will be made available for use by other researchers.

show abstract

Progressive Alignment and Discriminative Error Correction for Multiple OCR Engines

Cited by 20 publications

References 12 publications

Statistical learning for OCR error correction

Statistical learning for OCR error correction

Error Correction with In-domain Training across Multiple OCR System Outputs

A synthetic document image dataset for developing and evaluating historical document processing methods

Contact Info

Product

Resources

About