Typewritten OCR Model for Ethiopic Characters

被引:0
作者
Deneke, Bereket Siraw [1 ]
Aga, Rosa Tsegaye [1 ]
Samuel, Mesay [2 ]
Mulat, Abel [1 ]
Mulat, Ashenafi [1 ]
Abebe, Abel [1 ]
Mekonnen, Rahel [1 ]
Mulugeta, Hiwot [1 ]
Debelee, Taye Girma [1 ,3 ]
Gachena, Worku [1 ]
机构
[1] Ethiopian Artificial Intelligence Inst, Addis Ababa, Ethiopia
[2] Arbaminch Univ, Arbaminch, Ethiopia
[3] Addis Ababa Sci & Technol Univ, Dept Elect & Comp Engn, Addis Ababa, Ethiopia
来源
PAN-AFRICAN CONFERENCE ON ARTIFICIAL INTELLIGENCE, PT I, PANAFRICON AI 2023 | 2024年 / 2068卷
关键词
CNN; RNN; OCR; Tesseract; Transcribe; RECOGNITION;
D O I
10.1007/978-3-031-57624-9_14
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Optical Character Recognition (OCR) is the electronic conversion of images of computer-written, typewritten, handwritten, or printed text into machine-encoded text from a scanned document and a photo of a document. In Ethiopia, documents such as historical, office, and official documents have been documented in handwritten and type-written form, until recently. Thus, a large number of historical and essential documents are still in hard-copy form and at risk of disaster to be lost. Computer-written and handwritten OCR have been developed for different language characters including Ethiopian languages (Ethiopic characters), But not typewriter-written OCR for Ethiopic scripts. Like handwritten documents, large historical documents are typewritten documents in Ethiopia. Thus, the typewritten OCR is mandatory to preserve these documents. This study focuses on building an OCR model for typewritten documents that are written on Ethiopic characters. For the study, different Ethiopic characters have been collected from type-written documents, and 290 distinct characters have been segmented to construct augmented data to form various character variations and simulate the complexities encountered in real-world typewritten Amharic texts and enhance the adaptability of the OCR model. This technique aims to approximate the diversity inherent in the data. The model training framework leverages the capabilities of Tesseract, an open-source OCR engine, in conjunction with the artificially generated training set. The Tesseract's existing Amharic OCR model has been deployed as a base model, and the fine-tuning process has been adopted in a layered approach by employing 45,000 samples and spanning 4,800 iterations. The model has been evaluated using character error rate (CER). As per the evaluation, the model performed with 13% CER on the test set. For this study, the Tesseract model before fine-tuning and the Google Lense platform has been used as a baseline to evaluate the performance of the model. Accordingly, our model has outperformed both baselines by more than 10% margin.
引用
收藏
页码:250 / 261
页数:12
相关论文
共 18 条
  • [1] AHWR-Net: offline handwritten amharic word recognition using convolutional recurrent neural network
    Abdurahman, Fetulhak
    Sisay, Eyob
    Fante, Kinde Anlay
    [J]. SN APPLIED SCIENCES, 2021, 3 (08):
  • [2] Abish B, 2023, Nepali OCR
  • [3] Addis D, 2018, INT CONF SYST SCI EN
  • [4] Assabie Y, 2008, INT C PATT RECOG, P3043
  • [5] Belay B, 2019, IEEE IMAGE PROC, P2906, DOI [10.1109/icip.2019.8804407, 10.1109/ICIP.2019.8804407]
  • [6] Amharic character recognition using a fast signature based algorithm
    Cowell, J
    Hussain, F
    [J]. SEVENTH INTERNATIONAL CONFERENCE ON INFORMATION VISUALIZATION, PROCEEDINGS, 2003, : 384 - 389
  • [7] Easy OCR Converter, Free Online OCR in 100+ Languages. Free Convert Image to Text Online
  • [8] Gebremichael Hagos Tesfahun, 2022, Advances of Science and Technology: 9th EAI International Conference, ICAST 2021, Proceedings. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering (411), P559, DOI 10.1007/978-3-030-93709-6_38
  • [9] Multi-script handwritten digit recognition using multi-task learning
    Gondere, Mesay Samuel
    Schmidt-Thieme, Lars
    Sharma, Durga Prasad
    Scholz, Randolf
    [J]. JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2022, 43 (01) : 355 - 364
  • [10] i2OCR, i2OCR-Free Online OCR