scispace - formally typeset
Search or ask a question
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
More filters
Proceedings ArticleDOI
01 Sep 2016
TL;DR: This paper focuses on using Hidden Markov Model (HMM) based techniques to recognize the Named Entity (NE) for Gujarati language using HMM, which provides better performance and can be easily implemented for any languages.
Abstract: Named Entity Recognition (NER) is a method to search for a particular Named Entity (NE)[1] from a file or an image, recognize it and classify it into specified Entity Classes like Name, Location, Organization, Numbers and Others Categories. It is the most useful element of the technique known as Natural Language Processing (NLP) which makes text extraction very easy [2]. In this paper, we focus on using Hidden Markov Model (HMM) based techniques to recognize the Named Entity (NE) for Gujarati language. The main aim of using HMM is that it provides better performance and can be easily implemented for any languages. A remarkable amount of work has been carried out for many languages like English, Greek, Chinese etc. But, still a wide scope is open for Indian Origin Languages like Hindi, Gujarati, Devanagari etc. As Gujarati is not only the Indian Language, but a language that is most spoken in Gujarat. Thus, in this paper, we emphasis on proposing a NER based scheme for Gujarati Language using HMM.

4 citations

01 Jan 2010
TL;DR: A three tier strategy is suggested to recognize the handprinted characters of this script and the performance of the proposed scheme is reported in respect of recognition accuracy and time.
Abstract: Devanagari is an important script mostly used in South Asia and it is also the script of Hindi the mothertongue of majority of Indians. Though work done for the recognition of Devanagari hand-printed characters have been reported by some authors, but the availability of good ICR for this script is still a dream. In this paper, a three tier strategy is suggested to recognize the handprinted characters of this script. In primary and secondary stage classification, the structural properties of the script are exploited to avoid classification error. The results of all the three stages are reported on two classifiers i.e. MLP and SVM and the results achieved with the later are very good. The performance of the proposed scheme is reported in respect of recognition accuracy and time. The recognition rate achieved with the proposed scheme is 94.2% on our database consisting of more than 25000 characters belonging to 43 alphabets. The recognition rate has further improved to 95.3% when a conflict resolution strategy between some pair of characters is used.

4 citations

Journal ArticleDOI
TL;DR: The proposed model can be extended to recognize online handwritten words written in other script like Devanagari, Assamese, and Gurumukhi, etc and can be applied for offline word recognition purpose.
Abstract: This paper proposes a model which is used to recognize online handwritten Bangla words. This word recognition module comprises of different modules: preprocessing of the word samples, segmentation of words into basic strokes, recognizing the basic strokes using multilayer perceptron, followed by recognition of words using Hidden Markov Model (HMM) aided by Language Model (LM). For stroke recognition, two different feature extraction techniques (point-based and curvature-based procedures) are used using late fusion technique. Top 5 stroke recognition choices are used to construct HMM for the prediction of word sample. An N-gram LM is applied as a post-processing step to rectify the HMM outcomes if required. A total of 50 different word samples with 110 instances each are used to evaluate the proposed model. The overall stroke-level and word-level recognition accuracies obtained by this model are 95.4% and 90.3%, respectively. The proposed model can be extended to recognize online handwritten words written in other script like Devanagari, Assamese, and Gurumukhi, etc. The methodologies described in the manuscript can also be applied for offline word recognition purpose.

4 citations

Book ChapterDOI
01 Jan 2021
TL;DR: This paper gives a review of various techniques explored for Devanagari word/text and isolated character recognition in the past few years along with directions for future research.
Abstract: Optical character recognition techniques are capable of automatic translating of document images into equivalent character codes, so it helps in saving human energy as well as cost. These techniques can play a key role to improve or enhance the interaction between human and machine in many applications such as postal automation, signature verification, recognition of city names and automatic bank cheque processing/reading. This paper gives a review of various techniques explored for Devanagari word/text and isolated character recognition in the past few years. Different challenges to optical character recognition are also presented in this work. In the end, practical aspects towards the development of a robust optical character recognition system has been discussed along with directions for future research.

4 citations

Journal ArticleDOI
TL;DR: In this article , the authors proposed a new notion of handwritten numerals on the hypothesis that the handwritten characters are distinct deformations of the printed forms, which leads to easier recognition task with higher accuracy when superimposing handwritten numeral images onto the corresponding printed numeral image.

4 citations


Network Information
Related Topics (5)
Feature extraction
111.8K papers, 2.1M citations
77% related
Support vector machine
73.6K papers, 1.7M citations
75% related
Image segmentation
79.6K papers, 1.8M citations
74% related
Convolutional neural network
74.7K papers, 2M citations
74% related
Encryption
98.3K papers, 1.4M citations
73% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202342
202298
202148
202061
201938
201843