Handwritten Numeral Databases of Indian Scripts and Multistage Recognition of Mixed Numerals
01 Mar 2009-IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE Computer Society)-Vol. 31, Iss: 3, pp 444-457
TL;DR: P pioneering development of two databases for handwritten numerals of two most popular Indian scripts, a multistage cascaded recognition scheme using wavelet based multiresolution representations and multilayer perceptron classifiers and application for the recognition of mixed handwritten numeral recognition of three Indian scripts Devanagari, Bangla and English.
Abstract: This article primarily concerns the problem of isolated handwritten numeral recognition of major Indian scripts. The principal contributions presented here are (a) pioneering development of two databases for handwritten numerals of two most popular Indian scripts, (b) a multistage cascaded recognition scheme using wavelet based multiresolution representations and multilayer perceptron classifiers and (c) application of (b) for the recognition of mixed handwritten numerals of three Indian scripts Devanagari, Bangla and English. The present databases include respectively 22,556 and 23,392 handwritten isolated numeral samples of Devanagari and Bangla collected from real-life situations and these can be made available free of cost to researchers of other academic Institutions. In the proposed scheme, a numeral is subjected to three multilayer perceptron classifiers corresponding to three coarse-to-fine resolution levels in a cascaded manner. If rejection occurred even at the highest resolution, another multilayer perceptron is used as the final attempt to recognize the input numeral by combining the outputs of three classifiers of the previous stages. This scheme has been extended to the situation when the script of a document is not known a priori or the numerals written on a document belong to different scripts. Handwritten numerals in mixed scripts are frequently found in Indian postal mails and table-form documents.
Citations
More filters
05 Jun 2020
TL;DR: Experimental outcomes revealed that the proposed LSTM based method outperformed CNN with remarkable accuracy for the similar shaped numerals.
Abstract: In the promising era of Handwritten Numeral Recognition (HNR), despite Bangla being one of the major languages in the Indian subcontinent, fewer explorations have been done on Bangla numerals compared to other languages. Among the existing methods, several convolutional neural network (CNN) based method outperformed other methods. But CNN always gets confused with some specific Bangla numerals due to the similarity of shape and size of different numerals. The main purpose of this study is to expand Bangla HNR by considering a novel methodology with a Long Short-Term Memory (LSTM) network. In the proposed method, images are thinned and a sequence is extracted. These extracted sequences are used to classify using LSTM network. Both single-layer LSTM and Deep LSTM models are trained and performance tested on a benchmark dataset with a large number of samples. On the other hand, traditional CNN is also trained for better understanding. Experimental outcomes revealed that the proposed LSTM based method outperformed CNN with remarkable accuracy for the similar shaped numerals. Finally, the proposed method achieved a test set recognition rate of 98.03% which is better than or competitive to other prominent existing methods.
3 citations
Cites methods from "Handwritten Numeral Databases of In..."
...The CVPR, ISI dataset [11] [12] is used in this study which contains a total number of training and test samples are 19392 and 4000 written by 1106 individuals....
[...]
16 Dec 2016
TL;DR: Graph based features adequately model the cursivenes and are invariant to shape transformations and seem to be robust and resilient in this study.
Abstract: In this paper, the task of recognizing handwritten devanagari numerals by giving graph representation is introduced. Lipschitz embedding is explored to extract style, size invariant features from numeral graphs. Graph based features adequately model the cursivenes and are invariant to shape transformations. Recognition is carried out by SVM with radial basis function. Extensive experiments have been carried on standard dataset of CVPR ISI Kolkata. Comparative study of our results is presented with previous reported results on the dataset. From this study, graph representation seems to be robust and resilient.
3 citations
TL;DR: In this work an artificial neural network based classifier and statistical and structural method based feature extraction approach is used for the recognition of the script Devanagari.
Abstract: is the most effective way by which civilized people speaks. Devanagari is the basic Script widely used all over India. Many Indian languages like Hindi, Marathi, Rajasthani are based on Devanagari Script. Devanagari Scripts Hindi language is the third common language used all over the word. In the proposed work an artificial neural network based classifier and statistical and structural method based feature extraction approach is used for the recognition of the script. Optical isolated Marathi Characters are taken as an input image from the scanner. An input image is preprocessed and segmented. Features are extracted in terms of various structural and statistical features like End points, middle bar, loop, end bar, aspect ratio etc. Feature vector is applied to Self organizing map (SOM) which is one of the classifier of an artificial neural Network.SOM is trained for such 5000 different characters collected from 500 persons. The characters are classified into three different classes. The proposed classifier attains 93% accuracy.
3 citations
01 Oct 2016
TL;DR: This paper presents few techniques for optimizing the recognition accuracy at pre-classification stage, feature extraction stage and recognition stage of Devanagari script.
Abstract: Devanagari script has character set with rich structural features that makes the recognition of unconstrained handwritten Devanagari characters difficult However, these features can be used to divide the characters into different categories. This paper presents few techniques for optimizing the recognition accuracy at pre-classification stage, feature extraction stage and recognition stage. Initially, the pre-classification of the characters is done into different classes using various structural features. Then features are extracted using optimized feature extraction techniques. Finally, the recognition is done using neural network. In this paper, different neural networks are implemented and their performances are analyzed.
3 citations
Cites background from "Handwritten Numeral Databases of In..."
...Among different soft computing techniques, neural networks are found to be more popular for recognition of Indian numerals and characters, either printed or handwritten [13-18]....
[...]
01 Sep 2019
TL;DR: This work uses Conditional Generative Adversarial Networks (cGAN) for generating Class Based Character Generation and can help researchers to generate handwritten characters to enhance the perfomance of deep learning models.
Abstract: Bangla handwritten character recognition is known to be one of the most classical problem in the field of machine learning. In order to solve a machine learning problem one must thing is dataset. The more varied data a model sees the better it learns. Generative adversarial networks (GANs) are a group of neural networks that are used in unsupervised machine learning. It helps to resolve many difficult operations such as image generation from description, transforming low resolution image into high resolution, retrieving image contents given a small pattern etc. GAN's have many other promising applications in machine learning. There are many variations available for GAN. One of the variation of GAN is Conditional Generative Adversarial Networks(cGAN). This kind of GAN is used for generating a specific type of image. In this work we have used cGAN for generating Class Based Character Generation. This work can help researchers to generate handwritten characters to enhance the perfomance of deep learning models. We have trained this model to generate 50 Basic Bangla Characters, 10 Bangla Numerals and 24 Compound characters.
3 citations
Cites background or methods from "Handwritten Numeral Databases of In..."
...The dataset we used in this paper is BanglaLekhaIsolated[11], Indian Statistical Institue[12], CMATERDB[13]....
[...]
...Type CMATERDB1[13] ISI Dataset2 [12] BanglaLekhaIsolated Basic Character 15,103 30,966 98,950 Numerals 6,000 23,299 19,748 Compound Characters None None 47,407...
[...]
References
More filters
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.
42,067 citations
37,017 citations
Book•
16 Jul 1998
TL;DR: Thorough, well-organized, and completely up to date, this book examines all the important aspects of this emerging technology, including the learning process, back-propagation learning, radial-basis function networks, self-organizing systems, modular networks, temporal processing and neurodynamics, and VLSI implementation of neural networks.
Abstract: From the Publisher:
This book represents the most comprehensive treatment available of neural networks from an engineering perspective. Thorough, well-organized, and completely up to date, it examines all the important aspects of this emerging technology, including the learning process, back-propagation learning, radial-basis function networks, self-organizing systems, modular networks, temporal processing and neurodynamics, and VLSI implementation of neural networks. Written in a concise and fluid manner, by a foremost engineering textbook author, to make the material more accessible, this book is ideal for professional engineers and graduate students entering this exciting field. Computer experiments, problems, worked examples, a bibliography, photographs, and illustrations reinforce key concepts.
29,130 citations
TL;DR: In this paper, it is shown that the difference of information between the approximation of a signal at the resolutions 2/sup j+1/ and 2 /sup j/ (where j is an integer) can be extracted by decomposing this signal on a wavelet orthonormal basis of L/sup 2/(R/sup n/), the vector space of measurable, square-integrable n-dimensional functions.
Abstract: Multiresolution representations are effective for analyzing the information content of images. The properties of the operator which approximates a signal at a given resolution were studied. It is shown that the difference of information between the approximation of a signal at the resolutions 2/sup j+1/ and 2/sup j/ (where j is an integer) can be extracted by decomposing this signal on a wavelet orthonormal basis of L/sup 2/(R/sup n/), the vector space of measurable, square-integrable n-dimensional functions. In L/sup 2/(R), a wavelet orthonormal basis is a family of functions which is built by dilating and translating a unique function psi (x). This decomposition defines an orthogonal multiresolution representation called a wavelet representation. It is computed with a pyramidal algorithm based on convolutions with quadrature mirror filters. Wavelet representation lies between the spatial and Fourier domains. For images, the wavelet representation differentiates several spatial orientations. The application of this representation to data compression in image coding, texture discrimination and fractal analysis is discussed. >
20,028 citations
01 Jan 1988
TL;DR: This chapter contains sections titled: The Problem, The Generalized Delta Rule, Simulation Results, Some Further Generalizations, Conclusion.
Abstract: This chapter contains sections titled: The Problem, The Generalized Delta Rule, Simulation Results, Some Further Generalizations, Conclusion
17,604 citations
Additional excerpts
...Ç...
[...]