A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters
Hindi is the official language of India and used by a large population for several public services like postal, bank, judiciary, and public surveys. Efficient management of these services needs language-based automation. The proposed model addresses the problem of handwritten Hindi character recogni...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Future Internet |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-5903/13/9/239 |
| _version_ | 1797519220842430464 |
|---|---|
| author | Danveer Rajpal Akhil Ranjan Garg Om Prakash Mahela Hassan Haes Alhelou Pierluigi Siano |
| author_facet | Danveer Rajpal Akhil Ranjan Garg Om Prakash Mahela Hassan Haes Alhelou Pierluigi Siano |
| author_sort | Danveer Rajpal |
| collection | DOAJ |
| description | Hindi is the official language of India and used by a large population for several public services like postal, bank, judiciary, and public surveys. Efficient management of these services needs language-based automation. The proposed model addresses the problem of handwritten Hindi character recognition using a machine learning approach. The pre-trained DCNN models namely; InceptionV3-Net, VGG19-Net, and ResNet50 were used for the extraction of salient features from the characters’ images. A novel approach of fusion is adopted in the proposed work; the DCNN-based features are fused with the handcrafted features received from Bi-orthogonal discrete wavelet transform. The feature size was reduced by the Principal Component Analysis method. The hybrid features were examined with popular classifiers namely; Multi-Layer Perceptron (MLP) and Support Vector Machine (SVM). The recognition cost was reduced by 84.37%. The model achieved significant scores of precision, recall, and F1-measure—98.78%, 98.67%, and 98.69%—with overall recognition accuracy of 98.73%. |
| first_indexed | 2024-03-10T07:39:53Z |
| format | Article |
| id | doaj.art-a76411435f2a43038035bb1e13b24046 |
| institution | Directory Open Access Journal |
| issn | 1999-5903 |
| language | English |
| last_indexed | 2024-03-10T07:39:53Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Future Internet |
| spelling | doaj.art-a76411435f2a43038035bb1e13b240462023-11-22T13:10:28ZengMDPI AGFuture Internet1999-59032021-09-0113923910.3390/fi13090239A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi CharactersDanveer Rajpal0Akhil Ranjan Garg1Om Prakash Mahela2Hassan Haes Alhelou3Pierluigi Siano4Department of Electrical Engineering, Faculty of Engineering, J.N.V. University, Jodhpur 342001, IndiaDepartment of Electrical Engineering, Faculty of Engineering, J.N.V. University, Jodhpur 342001, IndiaPower System Planning Division, Rajasthan Rajya Vidyut Prasaran Nigam Ltd., Jaipur 302005, IndiaDepartment of Electrical Power Engineering, Tishreen University, Lattakia 2230, SyriaDepartment of Management & Innovation Systems, University of Salerno, 84084 Fisciano, ItalyHindi is the official language of India and used by a large population for several public services like postal, bank, judiciary, and public surveys. Efficient management of these services needs language-based automation. The proposed model addresses the problem of handwritten Hindi character recognition using a machine learning approach. The pre-trained DCNN models namely; InceptionV3-Net, VGG19-Net, and ResNet50 were used for the extraction of salient features from the characters’ images. A novel approach of fusion is adopted in the proposed work; the DCNN-based features are fused with the handcrafted features received from Bi-orthogonal discrete wavelet transform. The feature size was reduced by the Principal Component Analysis method. The hybrid features were examined with popular classifiers namely; Multi-Layer Perceptron (MLP) and Support Vector Machine (SVM). The recognition cost was reduced by 84.37%. The model achieved significant scores of precision, recall, and F1-measure—98.78%, 98.67%, and 98.69%—with overall recognition accuracy of 98.73%.https://www.mdpi.com/1999-5903/13/9/239bi-orthogonalDCNNDWTHindi charactershybrid-featuresfusion |
| spellingShingle | Danveer Rajpal Akhil Ranjan Garg Om Prakash Mahela Hassan Haes Alhelou Pierluigi Siano A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters Future Internet bi-orthogonal DCNN DWT Hindi characters hybrid-features fusion |
| title | A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters |
| title_full | A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters |
| title_fullStr | A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters |
| title_full_unstemmed | A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters |
| title_short | A Fusion-Based Hybrid-Feature Approach for Recognition of Unconstrained Offline Handwritten Hindi Characters |
| title_sort | fusion based hybrid feature approach for recognition of unconstrained offline handwritten hindi characters |
| topic | bi-orthogonal DCNN DWT Hindi characters hybrid-features fusion |
| url | https://www.mdpi.com/1999-5903/13/9/239 |
| work_keys_str_mv | AT danveerrajpal afusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT akhilranjangarg afusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT omprakashmahela afusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT hassanhaesalhelou afusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT pierluigisiano afusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT danveerrajpal fusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT akhilranjangarg fusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT omprakashmahela fusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT hassanhaesalhelou fusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters AT pierluigisiano fusionbasedhybridfeatureapproachforrecognitionofunconstrainedofflinehandwrittenhindicharacters |