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.

Sub-Stroke-Wise Relative Feature for Online Indic Handwriting Recognition

Abstract: The main problem of Bangla (Bengali) and Devanagari handwriting recognition is the shape similarity of characters. There are only a few pieces of work on writer-independent cursive online Indian text recognition, and the shape similarity problem needs more attention from the researchers. To handle the shape similarity problem of cursive characters of Bangla and Devanagari scripts, in this article, we propose a new category of features called ‘sub-stroke-wise relative feature’ (SRF) which are based on relative information of the constituent parts of the handwritten strokes. Relative information among some of the parts within a character can be a distinctive feature as it scales up small dissimilarities and enhances discrimination among similar-looking shapes. Also, contextual anticipatory phenomena are automatically modeled by this type of feature, as it takes into account the influence of previous and forthcoming strokes. We have tested popular state-of-the-art feature sets as well as proposed SRF using various (up to 20,000-word) lexicons and noticed that SRF significantly outperforms the state-of-the-art feature sets for online Bangla and Devanagari cursive word recognition.

The Formulation of Parameters for Type Design of Indian Script Based on Calligraphic Studies

Abstract: A number of parameters were formulated for better analysing the anatomy of Indic letterforms. This methodology contributes to the understanding of the intricacies of type design of complex Indian scripts.

A Robust OCR for Degraded Documents

Abstract: In the last two decades, many advances have been made in the field of document image analysis and recognition. In the recent past, several methods for recognizing Latin, Chinese, Japanese, and Arabic scripts have been proposed [7–9]. Until now, most of the OCR work has concentrated on high quality images and great success has been achieved by character recognition systems. Apart from these successes, there still exist two challenging problems in the field of recognition. The first one is optical character recognition (OCR) for low-quality images. Images having luminance variations, noise, and random degradation of text are difficult to read by OCR systems. The second open problem is that of recognizing off-line cursive handwritten character recognition [15]. Our work concentrates on the former one particularly for Devanagari script, which is the script for Hindi, Nepali, Marathi, and several other Indic languages. Together, these languages have a user base exceeding 500 million people. A great deal of effort has been made towards the development of OCR for Indian scripts [1–3, 10]. This chapter is concerned with the recognition of degraded Devanagri text documents. A major contribution in the area of Devanagari OCR is the Hindi OCR system developed by Chaudhari [2]. As remarked earlier, a majority of the above mentioned work has concentrated primarily on good quality document images and little work has been reported so far for the development of OCR for degraded Devanagari document images. Jawahar [4] presented a scheme based on SVM classifier, but this work was not focused on degraded document images. A major contribution towards the development of OCR for degraded documents is OCR for Chinese characters [13]. Apart from the advances that have been made towards

Line Parameter based Word-Level Indic Script Identification System

Abstract: In this paper, a line parameter based approach is presented to identify the handwritten scripts written in eight popular scripts. Since Optical Character Recognition OCR engines are usually script-dependent, automatic text recognition in multi-script environment requires a pre-processing module that helps identifying the scripts before processing the same through the respective OCR engine. The work becomes more challenging when it deals with handwritten document which is still a less explored research area. In this paper, a line parameter based approach is presented to identify the handwritten scripts written in eight popular scripts namely, Bangla, Devanagari, Gujarati, Gurumukhi, Manipuri, Oriya, Urdu, and Roman. A combination of Hough transform HT and Distance transform DT is used to extract the directional spatial features based on the line parameter. Experimentations are performed at word-level using multiple classifiers on a dataset of 12000 handwritten word images and Multi Layer Perceptron MLP classifier is found to be the best performing classifier showing an identification accuracy of 95.28%. The performance of the present technique is also compared with those of other state-of-the-art script identification methods on the same database. A combination of Hough transform HT and Distance transform DT is used to extract the directional spatial features based on the line parameter. Experimentation are performed at word-level on a total dataset of 12000 handwritten word images and Multi Layer Perceptron MLP classifier is found to be the best performing classifier showing an identification accuracy of 95.28%.

Zone-based keyword spotting in Bangla and Devanagari documents

Abstract: In this paper, we present a word spotting system in text lines for offline Indic scripts such as Bangla (Bengali) and Devanagari. Recently, it was shown that the zone-wise recognition method improves word recognition performance than the conventional full word recognition system in Indic scripts, like Bangla, Devanagari, Gurumukhi (Roy et al. in Pattern Recogn 60: 1057-1075, 26; Bhunia et al. in Pattern Recogn 79: 12–31, 6). Inspired from this idea we consider the zone segmentation approach and use middle zone information to improve the traditional word spotting performance. To avoid the problem of zone segmentation using heuristic approach, we propose here a new HMM based approach to segment the upper and lower zone components from the text line images. The candidate keywords are searched from a line without segmenting characters or words. Also, we propose a feature combining foreground and background information of text line images for keyword-spotting by character filler models. A significant improvement in performance is noted by using both foreground and background information instead of the individual one. Pyramid Histogram of Oriented Gradient (PHOG) feature has been used in our word spotting framework. From the experiment, it has been noted that the proposed zone-segmentation based system outperforms traditional approaches of word spotting.