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.

Component-level Script Classification Benchmark with CNN on AUTNT Dataset

Abstract: Script identification from camera images is a prerequisite for efficient end-to-end systems in multi-script environment. In recent times, due to the wide usage of digitized multi-lingual documents and images, efficient script identifier becomes an inevitable module in computer vision and pattern recognition applications. Here, a component-level multi-lingual identifier is designed based on CNN to set the benchmark performance on publicly available dataset named AUTNT. The model is evaluated using three different text scripts, viz., Bengali, Latin, and Devanagari. It yields reasonably high accuracy of 92.02% and 89.49% for document and scene component images, respectively, and 92.51% for overall text components irrespective of image sources. This result is first of its kind and it may be convincingly considered as a benchmark for component-level script classification from the said dataset.

Word Spotting for Indic Documents to Facilitate Retrieval

Abstract: With advances in the field of digitization of printed documents and several mass digitization projects underway, information retrieval and document search have emerged as key research areas. However, most of the current work in these areas is limited to English and a few oriental languages. The lack of efficient solutions for Indic scripts has hampered information extraction from a large body of documents of cultural and historical importance. This chapter presents two relevant topics in this area. First, we describe the use of a script-specific keyword spotting for Devanagari documents that makes use of domain knowledge of the script. Second, we address the needs of a digital library to provide access to a collection of documents from multiple scripts. This requires intelligent solutions which scale across different scripts. We present a script-independent keyword spotting approach for this purpose. Experimental results illustrate the efficacy of our methods.

Character Segmentation and Recognition of Indian Devanagari Script

Abstract: The paper presents Devanagari Character Segmentation and Recognition using neural networks. The hybrid features extraction technique which is the combination of geometric and statistical features is implemented. The geometric feature extraction technique uses directional features of Skeletonized Character image, whereas the statistical feature technique uses distribution of pixel density and Euclid features of the skeletonized character image. For classification, SVM (Support Vector Machine) and MLP (Multi Layer Perceptron) are used as classifiers. The Support Vector Machine has more accuracy as compare to MLP.

DS-P3SNet: An Efficient Classification Approach for Devanagari Script-Based P300 Speller Using Compact Channelwise Convolution and Knowledge Distillation

Abstract: Deep convolution neural network and its ensemble variant-based classification methods of P300 in the Devanagari script (DS)-based P300 speller (DS-P3S) have generated numerous training parameters. This is likely to increase the problems like computational complexity and overfitting. The recent attempts of researchers to overcome these problems are further deteriorating the accuracy due to the dense connectivity and channel-mix group convolution. Moreover, compressing the deep models in these attempts also found losing vital information. Therefore, to mitigate these problems, an efficient compact classification model called “DS-P3SNet” along with a knowledge distillation (KD) and transfer learning (TL) is proposed in this article. It includes: 1) extraction of rich morphological information across temporal region; 2) combination of channelwise and channel-mix-depthwise convolution (C2-DwCN) for efficient channel selection and extraction of spatial information with less number of trainable parameters; 3) channelwise convolution (Cw-CN) for classification to provide sparse connectivity; 4) knowledge distillation to reduce the tradeoff between accuracy and the number of trainable parameters; and 5) subject-to-subject transfer of learning to reduce subject variability. The trial-to-trial transfer of learning to reduce the tradeoff between the number of trials and accuracy. The experimentations were performed on a self-generated dataset of 20 words comprising of 79 DS characters collected from ten volunteer healthy subjects. An average accuracy of $95.32~{\pm }~0.85$ % and $94.64~{\pm }~0.68$ % were obtained for subject-dependent and subject-independent experiments, respectively. The trainable parameters were also reduced approximately by 2–34 times compared to existing models with improved or equivalent performance.