ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition
In recent years, Optical character recognition (OCR) has experienced a resurgence of interest especially for contemporary Arabic data. In fact, OCR development for printed and handwritten Arabic script is still a challenging task. These challenges are due to the specific characteristics of the Arabi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10476585/ |
| _version_ | 1797193517226786816 |
|---|---|
| author | Lamia Mosbah Ikram Moalla Tarek M. Hamdani Bilel Neji Taha Beyrouthy Adel M. Alimi |
| author_facet | Lamia Mosbah Ikram Moalla Tarek M. Hamdani Bilel Neji Taha Beyrouthy Adel M. Alimi |
| author_sort | Lamia Mosbah |
| collection | DOAJ |
| description | In recent years, Optical character recognition (OCR) has experienced a resurgence of interest especially for contemporary Arabic data. In fact, OCR development for printed and handwritten Arabic script is still a challenging task. These challenges are due to the specific characteristics of the Arabic script. In this work, we attempt to address these challenges by creating a deep learning OCR for Arabic document recognition called ADOCRNet. It is a novel deep learning framework whose architecture is built of layers of Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (BLSTM) trained using Connectionist Temporal Classification (CTC) algorithm. In order to assess the performance of our OCR, the proposed system is performed on two printed text datasets which are P-KHATT (text line images) and APTI (word images). It’s also evaluated on a handwritten Arabic text dataset IFN/ENIT (word images). According to the practical tests, the conceived model achieves strength recognition rates on the three datasets. ADOCRNet reaches a Character Error Rate (CER) of 0.01% on the P-KHATT dataset, 0.03% on the APTI dataset and a Word Error Rate (WER) of 1.09% on the IFN/ENIT dataset, which significantly outperforms the outcomes of the current systems. |
| first_indexed | 2024-04-24T05:41:38Z |
| format | Article |
| id | doaj.art-83ba6e1c8ae64ecd9a6c89df79ba41e6 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T05:41:38Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-83ba6e1c8ae64ecd9a6c89df79ba41e62024-04-23T23:00:20ZengIEEEIEEE Access2169-35362024-01-0112556205563110.1109/ACCESS.2024.337953010476585ADOCRNet: A Deep Learning OCR for Arabic Documents RecognitionLamia Mosbah0https://orcid.org/0009-0008-1285-5993Ikram Moalla1https://orcid.org/0000-0003-4703-3566Tarek M. Hamdani2https://orcid.org/0000-0002-8243-6056Bilel Neji3https://orcid.org/0000-0003-1147-4896Taha Beyrouthy4https://orcid.org/0000-0002-5939-7116Adel M. Alimi5REsearch Groups in Intelligent Machines (ReGIM-Lab), National Engineering School of Sfax (ENIS), University of Sfax, Sfax, TunisiaREsearch Groups in Intelligent Machines (ReGIM-Lab), National Engineering School of Sfax (ENIS), University of Sfax, Sfax, TunisiaREsearch Groups in Intelligent Machines (ReGIM-Lab), National Engineering School of Sfax (ENIS), University of Sfax, Sfax, TunisiaCollege of Engineering and Technology, American University of the Middle East, Egaila, KuwaitCollege of Engineering and Technology, American University of the Middle East, Egaila, KuwaitREsearch Groups in Intelligent Machines (ReGIM-Lab), National Engineering School of Sfax (ENIS), University of Sfax, Sfax, TunisiaIn recent years, Optical character recognition (OCR) has experienced a resurgence of interest especially for contemporary Arabic data. In fact, OCR development for printed and handwritten Arabic script is still a challenging task. These challenges are due to the specific characteristics of the Arabic script. In this work, we attempt to address these challenges by creating a deep learning OCR for Arabic document recognition called ADOCRNet. It is a novel deep learning framework whose architecture is built of layers of Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (BLSTM) trained using Connectionist Temporal Classification (CTC) algorithm. In order to assess the performance of our OCR, the proposed system is performed on two printed text datasets which are P-KHATT (text line images) and APTI (word images). It’s also evaluated on a handwritten Arabic text dataset IFN/ENIT (word images). According to the practical tests, the conceived model achieves strength recognition rates on the three datasets. ADOCRNet reaches a Character Error Rate (CER) of 0.01% on the P-KHATT dataset, 0.03% on the APTI dataset and a Word Error Rate (WER) of 1.09% on the IFN/ENIT dataset, which significantly outperforms the outcomes of the current systems.https://ieeexplore.ieee.org/document/10476585/Arabicdocument recognitionCNNsCTCdeep learningBLSTM |
| spellingShingle | Lamia Mosbah Ikram Moalla Tarek M. Hamdani Bilel Neji Taha Beyrouthy Adel M. Alimi ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition IEEE Access Arabic document recognition CNNs CTC deep learning BLSTM |
| title | ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition |
| title_full | ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition |
| title_fullStr | ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition |
| title_full_unstemmed | ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition |
| title_short | ADOCRNet: A Deep Learning OCR for Arabic Documents Recognition |
| title_sort | adocrnet a deep learning ocr for arabic documents recognition |
| topic | Arabic document recognition CNNs CTC deep learning BLSTM |
| url | https://ieeexplore.ieee.org/document/10476585/ |
| work_keys_str_mv | AT lamiamosbah adocrnetadeeplearningocrforarabicdocumentsrecognition AT ikrammoalla adocrnetadeeplearningocrforarabicdocumentsrecognition AT tarekmhamdani adocrnetadeeplearningocrforarabicdocumentsrecognition AT bilelneji adocrnetadeeplearningocrforarabicdocumentsrecognition AT tahabeyrouthy adocrnetadeeplearningocrforarabicdocumentsrecognition AT adelmalimi adocrnetadeeplearningocrforarabicdocumentsrecognition |