Impact of Ligature Coverage on Training Practical Urdu OCR Systems
Authors:
- Muhammad Ferjad NaeemMNMuhammad Ferjad Naeem
- Noor Ul Sehr ZiaNZNoor Ul Sehr Zia
- Aqsa Ahmed AwanAAAqsa Ahmed Awan
- Faisal ShafaitFSFaisal Shafait
- Adnan Ul HasanAHAdnan Ul Hasan
- 1 November 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 01, 131-136
Abstract
A major hurdle in the development of practical Urdu Nastaleeq script OCR is the lack of transcribed data, which is a pre-requisite for training machine learning algorithms. Most of the previous research has focused on UPTI, a publicly available data set with no particular focus on performance on real world images. UPTI contains only 6000 of the most probable 26,000 ligatures of Urdu. We build upon UPTI with a new data set, UPTI 2.0 that covers over 18,000 ligatures of Urdu Nastaleeq, hence covering over 70% of the ligatures that can practically occur. We further train a system on UPTI 2.0 and compare its performance against the only commercial Urdu Nastaleeq OCR system to date. Bidirectional Long Short-Term Memory (BDLSTM) network is employed with Connectionist Temporal Classification (CTC) layer as the recognizer. We show that systems trained on UPTI 2.0 outperform the commercial system.Keywords
This publication has 16 references indexed in Scilit:
- Urdu Nastaliq recognition using convolutional–recursive deep learningNeurocomputing, 2017
- Recognition of Urdu ligatures - a holistic approach2015
- The optical character recognition of Urdu-like cursive scriptsPattern Recognition, 2014
- Recognition of Nastalique Urdu ligatures2013
- Ligature Segmentation for Urdu OCR2013
- High-Performance OCR for Printed English and Fraktur Using LSTM Networks2013
- Context Sensitive Shape-Substitution in Nastaliq Writing System: Analysis and Formulation2007
- An Overview of the Tesseract OCR EngineNinth International Conference on Document Analysis and Recognition (ICDAR 2007), 2007
- Recognition of printed Urdu script2005
- Learning string-edit distanceIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998