Topic
Devanagari
About: Devanagari is a research topic. Over the lifetime, 655 publications have been published within this topic receiving 7428 citations. The topic is also known as: Deva nagari & Hindi Script.
Papers published on a yearly basis
Papers
More filters
••
TL;DR: A two-stage approach of deep learning is developed to enhance overall success of the proposed Devanagari Handwritten Character Recognition System (DHCRS), which requires very fewer trainable parameters and notably less training time to achieve state-of-the-art performance on a very small dataset.
Abstract: In order to rapidly build an automatic and precise system for image recognition and categorization, deep learning is a vital technology. Handwritten character classification also gaining more attention due to its major contribution in automation and specially to develop applications for helping visually impaired people. Here, the proposed work highlighting on fine-tuning approach and analysis of state-of-the-art Deep Convolutional Neural Network (DCNN) designed for Devanagari Handwritten characters classification. A new Devanagari handwritten characters dataset is generated which is publicly available. Datasets consist of total 5800 isolated images of 58 unique character classes: 12 vowels, 36 consonants and 10 numerals. In addition to this database, a two-stage VGG16 deep learning model is implemented to recognize those characters using two advanced adaptive gradient methods. A two-stage approach of deep learning is developed to enhance overall success of the proposed Devanagari Handwritten Character Recognition System (DHCRS). The first model achieves 94.84% testing accuracy with training loss of 0.18 on new dataset. Moreover, the second fine-tuned model requires very fewer trainable parameters and notably less training time to achieve state-of-the-art performance on a very small dataset. It achieves 96.55% testing accuracy with training loss of 0.12. We also tested the proposed model on four different benchmark datasets of isolated characters as well as digits of Indic scripts. For all the datasets tested, we achieved the promising results.
32 citations
••
01 Jan 2014TL;DR: From the experimental results, it has been found that the methodology provides higher recognition accuracies with lesser or equal numbers of features selected for each dataset.
Abstract: A new feature selection methodology on the basis of features’ combined class separability power, using the framework of Axiomatic Fuzzy Set (AFS) theory has been proposed here The AFS theory provides the rules for logic operations needed to interpret the combinations of features from the fuzzy feature set Based on these combinational rules, class separability power of the combined features is determined and subsequently the most powerful subset of the feature set is selected The performance of this methodology is evaluated upon for recognition of handwritten numerals of five popular Indic scripts viz Bangla, Devanagari, Roman, Telugu and Arabic with SVM based classifier using gradient based directional feature set and quad-tree based longest-run feature set separately and compared with six widely used feature selection techniques From the experimental results, it has been found that the methodology provides higher recognition accuracies with lesser or equal numbers of features selected for each dataset
32 citations
••
15 Dec 2009TL;DR: The objective of the current work is to recognize postal codes written in Roman, Devanagari, Bangla and Arabic scripts by using a script independent unified pattern classifier to classify any digit pattern of thesescripts into one of the 25 classes.
Abstract: The objective of the current work is to recognize postal codes written in Roman, Devanagari, Bangla and Arabic scripts. In the first stage 25 unique digit patterns are identified from the handwritten numeral patterns of the said four scripts. A script independent unified pattern classifier is then designed to classify any digit pattern of thesescripts into one of the 25 classes. In the next stage a rule-based script inference engine infers about the script of the numeric string, that invokes one of the four script specific classifiers. The average script-inference accuracy over a six digit numeric string is observed as 95.1% and the best recognition rates for the four script specific digit classifiers are obtained as 96.10%, 94.40%, 96.45 % and 95.60% respectively.
32 citations
••
24 Aug 2013TL;DR: A novel offline strategy for recognition of online handwritten Devanagari characters entered in an unconstrained manner, based on CNN, that allows writers to enter characters in any number or order of strokes and is also robust to certain amount of overwriting.
Abstract: In this paper, we introduce a novel offline strategy for recognition of online handwritten Devanagari characters entered in an unconstrained manner. Unlike the previous approaches based on standard classifiers - SVM, HMM, ANN and trained on statistical, structural or spectral features, our method, based on CNN, allows writers to enter characters in any number or order of strokes and is also robust to certain amount of overwriting. The CNN architecture supports an increased set of 42 Devanagari character classes. Experiments with 10 different configurations of CNN and for both Exponential Decay and Inverse Scale Annealing approaches to convergence, show highly promising results. In a further improvement, the final layer neuron outputs of top 3 configurations are averaged and used to make the classification decision, achieving an accuracy of 99.82% on the train data and 98.19% on the test data. This marks an improvement of 0.2% and 5.81%, for the train and test set respectively, over the existing state-of-the-art in unconstrained input. The data used for building the system is obtained from different parts of Devanagari writing states in India, in the form of isolated words. Character level data is extracted from the collected words using a hybrid approach and covers all possible variations owing to the different writing styles and varied parent word structures.
31 citations
••
TL;DR: A modified opposition-based multiobjective Harmony Search algorithm has been proposed to select the local regions from handwritten character images based on their rankings in a three-dimensional pareto-front based on recognition accuracy and redundancy.
31 citations