Tools for Semi-automatic Preparation of Training Data for OCR
作者： Ladislav Lenc , Jiří Martínek , Pavel Král
DOI: 10.1007/978-3-030-19823-7_29
关键词:
摘要: This work aims at data preparation for OCR systems based on recurrent neural networks. Precisely annotated are necessary training a network as well evaluation of methods. It is possible to synthesize the data, however such not that realistic real ones. Manual annotation thus still needed in many cases, especially case historical documents we focusing on. Although there several complex document processing, best our knowledge, simple tool completely missing. Therefore, propose and implement set tools utilizing artificial intelligence simplify process. These create ground truths line images used nowadays systems. Another contribution this paper making these freely available research purposes.
springer.com
sci-hub.st 参考文章(13)
Tapas Kanungo, Chang H. Lee, Jeff Czorapinski, Ivan Bella, TRUEVIZ: a groundtruth/metadata editing and visualizing toolkit for OCR document recognition and retrieval. ,vol. 4307, pp. 1- 12 ,(2000) , 10.1117/12.410825
C. Clausner, S. Pletschacher, A. Antonacopoulos, Aletheia - An Advanced Document Layout and Text Ground-Truthing System for Production Environments international conference on document analysis and recognition. pp. 48- 52 ,(2011) , 10.1109/ICDAR.2011.19
Thomas M. Breuel, The OCRopus open source OCR system document recognition and retrieval. ,vol. 6815, ,(2008) , 10.1117/12.783598
Kai Chen, Mathias Seuret, Hao Wei, Marcus Liwicki, Jean Hennebert, Rolf Ingold, None, Ground Truth Model, Tool, and Dataset for Layout Analysis of Historical Documents document recognition and retrieval. ,vol. 9402, pp. 940204- ,(2015) , 10.1117/12.2075858
Thomas M. Breuel, Adnan Ul-Hasan, Mayce Ali Al-Azawi, Faisal Shafait, High-Performance OCR for Printed English and Fraktur Using LSTM Networks international conference on document analysis and recognition. pp. 683- 687 ,(2013) , 10.1109/ICDAR.2013.140
Sepp Hochreiter, Jürgen Schmidhuber, Long short-term memory Neural Computation. ,vol. 9, pp. 1735- 1780 ,(1997) , 10.1162/NECO.1997.9.8.1735
Joost van Beusekom, Faisal Shafait, Thomas M. Breuel, Automated OCR Ground Truth Generation document analysis systems. pp. 111- 117 ,(2008) , 10.1109/DAS.2008.59
A. Graves, M. Liwicki, S. Fernandez, R. Bertolami, H. Bunke, J. Schmidhuber, A Novel Connectionist System for Unconstrained Handwriting Recognition IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 31, pp. 855- 868 ,(2009) , 10.1109/TPAMI.2008.137
Alex Graves, Santiago Fernández, Faustino Gomez, Jürgen Schmidhuber, Connectionist temporal classification Proceedings of the 23rd international conference on Machine learning - ICML '06. pp. 369- 376 ,(2006) , 10.1145/1143844.1143891
Alex Graves, Jürgen Schmidhuber, Offline Handwriting Recognition with Multidimensional Recurrent Neural Networks neural information processing systems. ,vol. 21, pp. 545- 552 ,(2008) , 10.1007/978-1-4471-4072-6_12