OCR with the Deep CNN Model for Ligature Script-Based Languages like Manchu
Authors:
Open Access
- 1 June 2021
- journal article
- research article
- Published by Wiley in Scientific Programming
- Vol. 2021, 1-9
Abstract
Manchu is a low-resource language that is rarely involved in text recognition technology. Because of the combination of typefaces, ordinary text recognition practice requires segmentation before recognition, which affects the recognition accuracy. In this paper, we propose a Manchu text recognition system divided into two parts: text recognition and text retrieval. First, a deep CNN model is used for text recognition, using a sliding window instead of manual segmentation. Second, text retrieval finds similarities within the image and locates the position of the recognized text in the database; this process is described in detail. We conducted comparative experiments on the FAST-NU dataset using different quantities of sample data, as well as comparisons with the latest model. The experiments revealed that the optimal results of the proposed deep CNN model reached 98.84%.Keywords
Funding Information
- Philosophy and Social Science Planning Project of Guangdong Province (GD19CYY13)
This publication has 18 references indexed in Scilit:
- Ligature based Urdu Nastaleeq sentence recognition using gated bidirectional long short term memoryCluster Computing, 2017
- Urdu Nastaliq recognition using convolutional–recursive deep learningNeurocomputing, 2017
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- Line and Ligature Segmentation of Urdu Nastaleeq TextIEEE Access, 2017
- Manchu character segmentation and recognition methodJournal of Discrete Mathematical Sciences and Cryptography, 2016
- Deep Residual Learning for Image Recognition2016
- Synthesis of Common Arabic Handwritings to Aid Optical Character Recognition ResearchSensors, 2016
- Robust Optical Recognition of Cursive Pashto Script Using Scale, Rotation and Location Invariant ApproachPLOS ONE, 2015
- Global Contrast Based Salient Region DetectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- A New Dataset Size Reduction Approach for PCA‐Based Classification in OCR ApplicationMathematical Problems in Engineering, 2014