Convolutional Autoencoder Based Feature Extraction and KNN Classifier for Handwritten MODI Script Character Recognition
- Title
- Convolutional Autoencoder Based Feature Extraction and KNN Classifier for Handwritten MODI Script Character Recognition
- Creator
- Joseph S.; George J.
- Description
- Character recognition is the process of identifying and classifying the images of printed or handwritten text and the conversion of that into machine-coded text. Deep learning techniques are efficiently used in the character recognition process. A Convolutional Autoencoder based technique for the character recognition of handwritten MODI script is proposed in this paper. MODI script was used for writing Marathi until the twentieth century. Though at present, Devnagari is taken over as the official script of Marathi, the historical importance of MODI script cannot be overlooked. MODI character recognition will not be an easy feat because of the various complexities of the script. Character recognition-related research of MODI script is in its initial stages. The proposed method is aimed to explore the use of a deep learning-based method for feature extraction and thereby building an efficient character recognition system for isolated handwritten MODI script. At the classification stage, the features extracted from the autoencoder are categorized using KNN classifier. Performance comparison of two different classifiers, such as KNN and SVM, is also carried out in this work. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
- Source
- Lecture Notes in Networks and Systems, Vol-290, pp. 142-149.
- Date
- 2021-01-01
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Subject
- Autoencoder; CNN; Handwritten character recognition; KNN; MODI script
- Coverage
- Joseph S., Science and Commerce for Women, Carmel College of Arts, Nuvem, Goa, India; George J., Christ University, Bangalore, India
- Rights
- Restricted Access
- Relation
- ISSN: 23673370; ISBN: 978-981164485-6
- Format
- Online
- Language
- English
- Type
- Conference paper
Collection
Citation
Joseph S.; George J., “Convolutional Autoencoder Based Feature Extraction and KNN Classifier for Handwritten MODI Script Character Recognition,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 25, 2025, https://archives.christuniversity.in/items/show/20575.