{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,15]],"date-time":"2026-08-15T17:46:51Z","timestamp":1786816011434,"version":"3.56.0"},"reference-count":27,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2020,5,2]],"date-time":"2020-05-02T00:00:00Z","timestamp":1588377600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100006595","name":"Unitatea Executiva pentru Finantarea Invatamantului Superior, a Cercetarii, Dezvoltarii si Inovarii","doi-asserted-by":"publisher","award":["PN-III-P1-1.2-PCCDI-2017-0689"],"award-info":[{"award-number":["PN-III-P1-1.2-PCCDI-2017-0689"]}],"id":[{"id":"10.13039\/501100006595","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Symmetry"],"abstract":"<jats:p>Optical Character Recognition (OCR) is the process of identifying and converting texts rendered in images using pixels to a more computer-friendly representation. The presented work aims to prove that the accuracy of the Tesseract 4.0 OCR engine can be further enhanced by employing convolution-based preprocessing using specific kernels. As Tesseract 4.0 has proven great performance when evaluated against a favorable input, its capability of properly detecting and identifying characters in more realistic, unfriendly images is questioned. The article proposes an adaptive image preprocessing step guided by a reinforcement learning model, which attempts to minimize the edit distance between the recognized text and the ground truth. It is shown that this approach can boost the character-level accuracy of Tesseract 4.0 from 0.134 to 0.616 (+359% relative change) and the F1 score from 0.163 to 0.729 (+347% relative change) on a dataset that is considered challenging by its authors.<\/jats:p>","DOI":"10.3390\/sym12050715","type":"journal-article","created":{"date-parts":[[2020,5,5]],"date-time":"2020-05-05T06:41:20Z","timestamp":1588660880000},"page":"715","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":60,"title":["Improving the Accuracy of Tesseract 4.0 OCR Engine Using Convolution-Based Preprocessing"],"prefix":"10.3390","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7777-1928","authenticated-orcid":false,"given":"Dan","family":"Sporici","sequence":"first","affiliation":[{"name":"Computer Science and Engineering Department, Faculty of Automatic Control and Computers, Politehnica University of Bucharest, Splaiul Independen\u021bei 313, 060042 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Elena","family":"Cu\u0219nir","sequence":"additional","affiliation":[{"name":"Computer Science and Engineering Department, Faculty of Automatic Control and Computers, Politehnica University of Bucharest, Splaiul Independen\u021bei 313, 060042 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2987-4022","authenticated-orcid":false,"given":"Costin-Anton","family":"Boiangiu","sequence":"additional","affiliation":[{"name":"Computer Science and Engineering Department, Faculty of Automatic Control and Computers, Politehnica University of Bucharest, Splaiul Independen\u021bei 313, 060042 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2020,5,2]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Smith, R. (2007, January 23\u201326). An overview of the Tesseract OCR engine. Proceedings of the Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Parana, Brazil.","DOI":"10.1109\/ICDAR.2007.4376991"},{"key":"ref_2","first-page":"50","article-title":"Optical character recognition by open source OCR tool tesseract: A case study","volume":"55","author":"Patel","year":"2012","journal-title":"Int. J. Comput. Appl."},{"key":"ref_3","unstructured":"(2020, January 23). TESSERACT(1) Manual Page. Available online: https:\/\/github.com\/tesseract-ocr\/tesseract\/blob\/master\/doc\/tesseract.1.asc."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Sporici, D., Chiroiu, M., and Cioc\u00eerlan, D. (2018, January 8\u20139). An Evaluation of OCR Systems Against Adversarial Machine Learning. Proceedings of the International Conference on Security for Information Technology and Communications, SECITC 2018, Bucharest, Romania.","DOI":"10.1007\/978-3-030-12942-2_11"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Harraj, A.E., and Naoufal, R. (2015). OCR Accuracy Improvement on Document Images through a Novel Pre-Processing Approach. arXiv.","DOI":"10.5121\/sipij.2015.6401"},{"key":"ref_6","first-page":"10","article-title":"Unsharp Masking","volume":"16","author":"Malin","year":"1977","journal-title":"Aas Photo Bull."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"62","DOI":"10.1109\/TSMC.1979.4310076","article-title":"A threshold selection method from gray-level histograms","volume":"9","author":"Otsu","year":"1979","journal-title":"Ieee Trans. Syst. ManCybern."},{"key":"ref_8","unstructured":"Koistinen, M., Kettunen, K., and Kervinen, J. (2018). How to Improve Optical Character Recognition of Historical Finnish Newspapers Using Open Source Tesseract OCR Engine, Adam Mickiewicz University."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1037","DOI":"10.1109\/ICPR.2002.1048482","article-title":"Text localization, enhancement and binarization in multimedia documents","volume":"Volume 2","author":"Wolf","year":"2002","journal-title":"Object Recognition Supported by User Interaction for Service Robots"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"355","DOI":"10.1016\/S0734-189X(87)80186-X","article-title":"Adaptive histogram equalization and its variations","volume":"39","author":"Pizer","year":"1987","journal-title":"Comput. Vis. Graph. Image Process."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Breuel, T.M. (2007, January 23\u201326). The hOCR microformat for OCR workflow and results. Proceedings of the Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Parana, Brazil.","DOI":"10.1109\/ICDAR.2007.4377078"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Reul, C., Springmann, U., Wick, C., and Puppe, F. (2018, January 24\u201327). Improving OCR accuracy on early printed books by utilizing cross fold training and voting. Proceedings of the 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), Vienna, Austria.","DOI":"10.1109\/DAS.2018.30"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Bui, Q.A., Mollard, D., and Tabbone, S. (2017, January 9\u201315). Selecting automatically pre-processing methods to improve OCR performances. Proceedings of the 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan.","DOI":"10.1109\/ICDAR.2017.36"},{"key":"ref_14","unstructured":"Shen, M., and Lei, H. (2015, January 15\u201317). Improving OCR performance with background image elimination. Proceedings of the 2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), Zhangjiajie, China."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Thompson, P., McNaught, J., and Ananiadou, S. (October, January 28). Customised OCR correction for historical medical text. Proceedings of the 2015 Digital Heritage, Granada, Spain.","DOI":"10.1109\/DigitalHeritage.2015.7413829"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"de Jager, C., and Nel, M. (2019). Business Process Automation: A Workflow Incorporating Optical Character Recognition and Approximate String and Pattern Matching for Solving Practical Industry Problems. Appl. Syst. Innov., 2.","DOI":"10.3390\/asi2040033"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Ford, G., Hauser, S.E., Le, D.X., and Thoma, G.R. (2001, January 24\u201325). Pattern matching techniques for correcting low-confidence OCR words in a known context. Proceedings of the Document Recognition and Retrieval VIII, San Jose, CA, USA.","DOI":"10.1117\/12.410842"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Priambada, S., and Widyantoro, D.H. (2017, January 1\u20133). Levensthein distance as a post-process to improve the performance of OCR in written road signs. Proceedings of the 2017 Second International Conference on Informatics and Computing (ICIC), Jayapura, Indonesia.","DOI":"10.1109\/IAC.2017.8280534"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Brisinello, M., Grbi\u0107, R., Pul, M., and An\u0111eli\u0107, T. (2017, January 18\u201320). Improving optical character recognition performance for low quality images. Proceedings of the 2017 International Symposium ELMAR, Zadar, Croatia.","DOI":"10.23919\/ELMAR.2017.8124460"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Ki\u0161\u0161, M., Hradi\u0161, M., and Kodym, O. (2019). Brno Mobile OCR Dataset. arXiv.","DOI":"10.1109\/ICDAR.2019.00218"},{"key":"ref_21","unstructured":"Kurakin, A., Goodfellow, I., and Bengio, S. (2017, January 24\u201326). Adversarial machine learning at scale. Proceedings of the 2017 International Conference on Learning Representations (ICLR), Toulon, France."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Rice, S.V., Kanai, J., and Nartker, T.A. (1994). An algorithm for matching OCR-generated text strings. Document Image Analysis, World Scientific.","DOI":"10.1142\/9789812797797_0015"},{"key":"ref_23","unstructured":"(2020, January 23). International Workshop on Document Analysis Systems (DAS) 2016, Chapter 7. Available online: https:\/\/github.com\/tesseract-ocr\/docs\/blob\/master\/das_tutorial2016."},{"key":"ref_24","unstructured":"Fujimoto, S., van Hoof, H., and Meger, D. (2018). Addressing function approximation error in actor-critic methods. arXiv."},{"key":"ref_25","unstructured":"Cobbe, K., Klimov, O., Hesse, C., Kim, T., and Schulman, J. (2018). Quantifying generalization in reinforcement learning. arXiv."},{"key":"ref_26","first-page":"470","article-title":"Voting-Based OCR System","volume":"10","author":"Boiangiu","year":"2016","journal-title":"Proc. J. ISOM J. Inf. Syst. Oper. Manag."},{"key":"ref_27","first-page":"10","article-title":"Automatic Correction of OCR Results Using Similarity Detection for Words and Fonts","volume":"10","author":"Boiangiu","year":"2016","journal-title":"Int. J. Appl. Math. Inform."}],"container-title":["Symmetry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-8994\/12\/5\/715\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,13]],"date-time":"2025-10-13T13:32:28Z","timestamp":1760362348000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-8994\/12\/5\/715"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,5,2]]},"references-count":27,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2020,5]]}},"alternative-id":["sym12050715"],"URL":"https:\/\/doi.org\/10.3390\/sym12050715","relation":{},"ISSN":["2073-8994"],"issn-type":[{"value":"2073-8994","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,5,2]]}}}