A Multi-Layer Holistic Approach for Cursive Text Recognition
Urdu is a widely spoken and narrated language in several South-Asian countries and communities worldwide. It is relatively hard to recognize Urdu text compared to other languages due to its cursive writing style. The Urdu text script belongs to a non-Latin cursive family script like Arabic, Hindi an...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/24/12652 |
| _version_ | 1797461615073820672 |
|---|---|
| author | Muhammad Umair Muhammad Zubair Farhan Dawood Sarim Ashfaq Muhammad Shahid Bhatti Mohammad Hijji Abid Sohail |
| author_facet | Muhammad Umair Muhammad Zubair Farhan Dawood Sarim Ashfaq Muhammad Shahid Bhatti Mohammad Hijji Abid Sohail |
| author_sort | Muhammad Umair |
| collection | DOAJ |
| description | Urdu is a widely spoken and narrated language in several South-Asian countries and communities worldwide. It is relatively hard to recognize Urdu text compared to other languages due to its cursive writing style. The Urdu text script belongs to a non-Latin cursive family script like Arabic, Hindi and Chinese. Urdu is written in several writing styles, among which ‘Nastaleeq’ is the most popular and widely used font style. A gap still poses a challenge for localization/detection and recognition of Urdu Nastaleeq text as it follows modified version of Arabic script. This research study presents a methodology to recognize and classify Urdu text in Nastaleeq font, regardless of the text position in the image. The proposed solution is comprised of a two-step methodology. In the first step, text detection is performed using the Connected Component Analysis (CCA) and Long Short-Term Memory Neural Network (LSTM). In the second step, a hybrid Convolution Neural Network and Recurrent Neural Network (CNN-RNN) architecture is deployed to recognize the detected text. The image containing Urdu text is binarized and segmented to produce a single-line text image fed to the hybrid CNN-RNN model, which recognizes the text and saves it in a text file. The proposed technique outperforms the existing ones by achieving an overall accuracy of 97.47%. |
| first_indexed | 2024-03-09T17:22:53Z |
| format | Article |
| id | doaj.art-441c66529ef648bfbb70043ee0c99f5d |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T17:22:53Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-441c66529ef648bfbb70043ee0c99f5d2023-11-24T13:02:23ZengMDPI AGApplied Sciences2076-34172022-12-0112241265210.3390/app122412652A Multi-Layer Holistic Approach for Cursive Text RecognitionMuhammad Umair0Muhammad Zubair1Farhan Dawood2Sarim Ashfaq3Muhammad Shahid Bhatti4Mohammad Hijji5Abid Sohail6Faculty of Information Technology & Computer Science, University of Central Punjab, Lahore 54000, PakistanFaculty of Information Technology & Computer Science, University of Central Punjab, Lahore 54000, PakistanFaculty of Information Technology & Computer Science, University of Central Punjab, Lahore 54000, PakistanFaculty of Information Technology & Computer Science, University of Central Punjab, Lahore 54000, PakistanFaculty of Information Technology & Computer Science, University of Central Punjab, Lahore 54000, PakistanFaculty of Computers and Information Technology, University of Tabuk, Tabuk 47921, Saudi ArabiaDepartment of Computer Science, COMSATS University Islamabad, Lahore Campus, Lahore 54000, Pakistan Urdu is a widely spoken and narrated language in several South-Asian countries and communities worldwide. It is relatively hard to recognize Urdu text compared to other languages due to its cursive writing style. The Urdu text script belongs to a non-Latin cursive family script like Arabic, Hindi and Chinese. Urdu is written in several writing styles, among which ‘Nastaleeq’ is the most popular and widely used font style. A gap still poses a challenge for localization/detection and recognition of Urdu Nastaleeq text as it follows modified version of Arabic script. This research study presents a methodology to recognize and classify Urdu text in Nastaleeq font, regardless of the text position in the image. The proposed solution is comprised of a two-step methodology. In the first step, text detection is performed using the Connected Component Analysis (CCA) and Long Short-Term Memory Neural Network (LSTM). In the second step, a hybrid Convolution Neural Network and Recurrent Neural Network (CNN-RNN) architecture is deployed to recognize the detected text. The image containing Urdu text is binarized and segmented to produce a single-line text image fed to the hybrid CNN-RNN model, which recognizes the text and saves it in a text file. The proposed technique outperforms the existing ones by achieving an overall accuracy of 97.47%.https://www.mdpi.com/2076-3417/12/24/12652text detectiontext recognitionnatural language processingnatural language understandingmachine learningdeep learning applications |
| spellingShingle | Muhammad Umair Muhammad Zubair Farhan Dawood Sarim Ashfaq Muhammad Shahid Bhatti Mohammad Hijji Abid Sohail A Multi-Layer Holistic Approach for Cursive Text Recognition Applied Sciences text detection text recognition natural language processing natural language understanding machine learning deep learning applications |
| title | A Multi-Layer Holistic Approach for Cursive Text Recognition |
| title_full | A Multi-Layer Holistic Approach for Cursive Text Recognition |
| title_fullStr | A Multi-Layer Holistic Approach for Cursive Text Recognition |
| title_full_unstemmed | A Multi-Layer Holistic Approach for Cursive Text Recognition |
| title_short | A Multi-Layer Holistic Approach for Cursive Text Recognition |
| title_sort | multi layer holistic approach for cursive text recognition |
| topic | text detection text recognition natural language processing natural language understanding machine learning deep learning applications |
| url | https://www.mdpi.com/2076-3417/12/24/12652 |
| work_keys_str_mv | AT muhammadumair amultilayerholisticapproachforcursivetextrecognition AT muhammadzubair amultilayerholisticapproachforcursivetextrecognition AT farhandawood amultilayerholisticapproachforcursivetextrecognition AT sarimashfaq amultilayerholisticapproachforcursivetextrecognition AT muhammadshahidbhatti amultilayerholisticapproachforcursivetextrecognition AT mohammadhijji amultilayerholisticapproachforcursivetextrecognition AT abidsohail amultilayerholisticapproachforcursivetextrecognition AT muhammadumair multilayerholisticapproachforcursivetextrecognition AT muhammadzubair multilayerholisticapproachforcursivetextrecognition AT farhandawood multilayerholisticapproachforcursivetextrecognition AT sarimashfaq multilayerholisticapproachforcursivetextrecognition AT muhammadshahidbhatti multilayerholisticapproachforcursivetextrecognition AT mohammadhijji multilayerholisticapproachforcursivetextrecognition AT abidsohail multilayerholisticapproachforcursivetextrecognition |