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.

A Convolutional Autoencoder based Keyword Spotting in Historical Handwritten Devanagari Documents

Abstract: There exist huge amount of valuable historical Devanagari documents archived in many national libraries that need to be preserved in digital form. Additionally, there is a growing requirement in the area of document image processing for automation and information extraction from old handwritten documents. Retrieval of relevant information from handwritten historical document images would ideally necessitate an efficient An alternate approach to accurate document transcription is to use a keyword spotting method. The keyword spotting system's primary application is to create digital libraries that facilitate quick searching and browsing of old manuscripts in order to preserve the world's cultural heritage. For word spotting in handwritten manuscripts, we propose a convolution neural network (CNN) architecture based on Autoencoder representation. The proposed technique was demonstrated using historical handwritten Devanagari manuscript which is collected from Oriental Research Institute of University of Mysore, Mysore. As a result, the proposed convolution neural network (CNN) method exhibits superior accuracy with favorable results.

Recent Advancement In Devanagari Hand Writing Recognition

Abstract: A technical Review of the state of art: techniques in Devanagari hand writing recognition is presented. The handwriting recognition is matured for Roman, Japanese and Chinese and Arabian language scripts but for Indian languages a lot of scope is there. For Indian languages most of the work is limited to isolated characters and numerals. Compound characters and word recognition has not been explored to that extent.

Script Recognition using GLCM and DWT Features

Abstract: In this paper a method is proposed for identification of Roman, Devanagari, Kannada, Tamil, Telugu and Malayalam scripts at text block level using features of Correlation property of Gray Level Co-occurrence Matrix (GLCM) and multi resolutionality of Discrete Wavelet Transform (DWT) of input handwritten document text blocks. The two-dimensional DWT extracts spatial features and Correlation of GLCM is used to extract texture features. Typically it can be observed that the patterns of any handwritten text block encompass spatial texture primitives. Therefore, the primary aim of this paper is to show the efficiency of DWT and Correlation of GLCM in describing the handwritten text blocks of six Indian scripts. Exhaustive experimentations were conducted on a dataset of 100 text blocks of each script, with bi-script and tri-script combinations of six scripts and script recognition is carried out using three classifiers namely nearest neighbor (NN), LDA and SVM. Using SVM classifier average script classification accuracy achieved in case of bi-script and tri-script combinations are 96.4333% and 93.9833% respectively.

Detection of Devanagari Text from Wild Images Through Image Processing Techniques

Abstract: Multilingual society makes it difficult for a common man to understand the text written in various languages. Being the national language of India, Hindi is widely understood by the masses. Detecting text from randomly available images and translating it into a local choice of language is one of the peculiar and less explored research areas in image processing and is very popular among the scientific community. Moreover, it is useful for removing the language barrier present on earth. In this paper, we propose a complete solution to detect text from natural scene images as in wild images and converting it into a textual format. The proposed work follows a combined approach of stroke width transform (SWT) and maximally stable extremal regions (MSER) to reduce the computational time. There are three steps involved in our proposed approach. In the first step, we are first segmenting out the text by using the binarization and SWT algorithm. Then, the second step implements the MSER to detect the text regions and bounding boxes. The regions detected in the second step acts as letter candidates, which are then recognized in the third step using the optical character recognition approach (OCR). The experimental results show that the characters are being recognized at an accuracy of 77% approximately, outperforming the existing approaches in the literature.