Abstract:An optical character recognition (OCR) system segments the character from the given document before recognizing it. The recognition of such character images requires the class labels to be associated with each character sample in the training set, and this requires the placing of all the samples of each segmented character in various distinct folders. However, it has to be done manually, and thus, it is a time-consuming process. The ancient documents suffer from humidity spots, ink stains, and faded portions of text which makes the character recognition task even more challenging for the ancient documents. The present article proposes a novel semi-self-supervised learning-based OCR method to recognize each character segmented from the ancient documents handwritten in Devanagari and Maithili scripts. The proposed method has two modules—feature extraction module and recognition module. The feature extraction module has extracted deep hierarchical features from each pre-segmented character image employing generative self-supervised learning approach. The recognition module has focused on important features using an attention mechanism and learns the long temporal sequence using the Gated Recurrent Unit variant of recurrent neural network classifier to classify each segmented character into its proper class. The feature extraction module in the proposed method has been trained using the 60% (unlabelled) of the dataset, whereas the recognition module has been trained using the 5% (manually labelled) of the dataset. The performance of the proposed novel OCR method has been evaluated on two self-generated datasets of ancient handwritten documents in Devanagari and Maithili scripts. The experimental results demonstrate that the proposed OCR method outperforms the state-of-the-art (SOTA) methods in this regard. The proposed OCR method has improved the character recognition accuracy in comparison with the SOTA methods by 2.27% and 3.48% in Devanagari and Maithili scripts, respectively.

Image Based Character Recognition, Documentation System To Decode Inscription From Temple

A Novel Approach to OCR using Image Recognition based Classification for Ancient Tamil Inscriptions in Temples

A Character Recognition Scheme Based on Object Oriented Design for Tibetan Buddhist Texts.

A semi-self-supervised learning model to recognize handwritten characters in ancient documents in Indian scripts

CNN-Bidirectional LSTM Based Optical Character Recognition of Sanskrit Manuscripts : A Comprehensive Systematic Literature Review

Adapting the Tesseract Open-Source OCR Engine for Tamil and Sinhala Legacy Fonts and Creating a Parallel Corpus for Tamil-Sinhala-English

Optical Character Recognition using Convolutional Neural Networks for Ashokan Brahmi Inscriptions

Handwritten OCR for Indic Scripts: A Comprehensive Overview of Machine Learning and Deep Learning Techniques

Advanced Digital Image Processing Technique based Optical Character Recognition of Scanned Document

Design And Development Of An Ancient Chinese Document Recognition System

Study of Tesseract OCR

Indian script character recognition: a survey

Handwritten Text Recognition Using Convolutional Neural Network

Performance of Binarization Algorithms on Tamizhi Inscription Images: An Analysis

Hilditchs Algorithm Based Tamil Character Recognition

Handwritten Tamil Character Recognition UsingDeep Learning

South Indian Character Recognition Using Statistical Feature Extraction and Distance Classifier

uTHCD: A New Benchmarking for Tamil Handwritten OCR

Handwritten Character Recognition of South Indian Scripts: A Review

Handwritten Digit Recognition Using Improved Bounding Box Recognition Technique

Recognition of Handwritten Textual Annotations using Tesseract Open Source OCR Engine for information Just In Time (iJIT)