Handwritten Character Recognition Using KNN and SVM Based Classifier over Feature Vector from Autoencoder

Mahapatra, Dibyakanti; Choudhury, Chandrajit; Karsh, Ram Kumar

doi:10.1007/978-981-15-6315-7_25

Cited by 9 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Four handwritten character recognition methods were studied by [28], achieving high accuracy rates of 89.02%, 86.67%, and 95.3%, respectively, contrasted with cutting-edge techniques.…”

Section: B) K Nearest Neighbor Classifiermentioning

confidence: 99%

Handwritten OCR for Indic Scripts: A Comprehensive Overview of Machine Learning and Deep Learning Techniques

Shaik Moinuddin Ahmed

2023

IJRITCC

View full text Add to dashboard Cite

The potential uses of cursive optical character recognition, commonly known as OCR, in a number of industries, particularly document digitization, archiving, even language preservation, have attracted a lot of interest lately. In the framework of optical character recognition (OCR), the goal of this research is to provide a thorough understanding of both cutting-edge methods and the unique difficulties presented by Indic scripts. A thorough literature search was conducted in order to conduct this study, during which time relevant publications, conference proceedings, and scientific files were looked for up to the year 2023. As a consequence of the inclusion criteria that were developed to concentrate on studies only addressing Handwritten OCR on Indic scripts, 53 research publications were chosen as the process's outcome. The review provides a thorough analysis of the methodology and approaches employed in the chosen study. Deep neural networks, conventional feature-based methods, machine learning techniques, and hybrid systems have all been investigated as viable answers to the problem of effectively deciphering Indian scripts, because they are famously challenging to write. To operate, these systems require pre-processing techniques, segmentation schemes, and language models. The outcomes of this methodical examination demonstrate that despite the fact that Hand Scanning for Indic script has advanced significantly, room still exists for advancement. Future research could focus on developing trustworthy models that can handle a range of writing styles and enhance accuracy using less-studied Indic scripts. This profession may advance with the creation of collected datasets and defined standards.

show abstract

“…Four handwritten character recognition methods were studied by [28], achieving high accuracy rates of 89.02%, 86.67%, and 95.3%, respectively, contrasted with cutting-edge techniques.…”

Section: B) K Nearest Neighbor Classifiermentioning

confidence: 99%

Handwritten OCR for Indic Scripts: A Comprehensive Overview of Machine Learning and Deep Learning Techniques

Shaik Moinuddin Ahmed

2023

IJRITCC

View full text Add to dashboard Cite

show abstract

“…Quadratic discriminant function (QDF), KNN, modified QDF, linear discriminant function (LDF), mixture of Gaussian, regularized discriminant analysis (RDA), and mixture of LDF are a few of Gaussian rules-based statistical methods which are widely used in OCR. [50][51][52][53] Urdu OCRs: Urdu character recognition remains a challenging problem due to the complexity of Arabic script which is cursive in nature. In a research segmentation of Urdu characters is proposed on the basis of complexity level, that is, complex level, semi-complex level, and simple level.…”

Section: Ta B L E 1 a Comparison Between The Existing Techniques Of O...mentioning

confidence: 99%

“…Some statistical methods are based on Bayesian rules where two approaches, that is, parametric based and nonparametric are in practice. Quadratic discriminant function (QDF), KNN, modified QDF, linear discriminant function (LDF), mixture of Gaussian, regularized discriminant analysis (RDA), and mixture of LDF are a few of Gaussian rules‐based statistical methods which are widely used in OCR 50–53 …”

Section: Related Workmentioning

confidence: 99%

Meta‐feature based few‐shot Siamese learning for Urdu optical character recognition

Naseer

Zafar

2022

Computational Intelligence

View full text Add to dashboard Cite

Standard convolution neural network (CNN) achieves high level of accuracy for the recognition of characters in different languages. However, like other deep neural networks, training of CNN requires a substantial amount of data. Lack of sufficient training data invokes dataset bias, during learning process, which leads to a decay in the performance of CNN. The limitation of training data can be addressed by using few-shot learners. In this research, CNN-based few-shot Siamese learner is trained on meta-features, extracted from Urdu text images using a novel graph-based normal to tangent line (GNTL) technique, for Urdu optical character recognition (OCR) across different font sizes. The learner is trained on three corpora (datasets) including one benchmark corpus "Centre for Language Engineering Text Images" and two other corpora, that is, "Urdu Thickness Graphs" (UTG) and "Urdu OCR Font 16 to 36" (UOF) which are developed and released in this research. 80% of data is used for training while 20% of data is used for testing. To create UTG corpus, the proposed novel feature extraction technique GNTL is used and a meta-features-based corpus is developed in form of thickness graphs. The third corpus UOF is based on five different font sizes, that is, 16, 20, 26, 30, and 36. The performance of few-shot Siamese learner is compared with a standard CNN, trained on the same three corpora. Meta-feature based few-shot Siamese learner achieves a promising recognition accuracy and outperforms

show abstract

“…Mahapatra et al [15] applied learning viz auto encoders to classify handwritten Kannada characters using KNN and SVM classifiers. Later, capsule networks for recognition of low resource languages such as Kannada is investigated by Abeysinghe and Perera [16].…”

Section: Introductionmentioning

confidence: 99%

Robust recognition technique for handwritten Kannada character recognition using capsule networks

Rani

Manohar²,

Hariprasad³

et al. 2022

IJECE

View full text Add to dashboard Cite

<p>Automated reading of handwritten Kannada documents is highly challenging due to the presence of vowels, consonants and its modifiers. The variable nature of handwriting styles aggravates the complexity of machine based reading of handwritten vowels and consonants. In this paper, our investigation is inclined towards design of a deep convolution network with capsule and routing layers to efficiently recognize Kannada handwritten characters. Capsule network architecture is built of an input layer, two convolution layers, primary capsule, routing capsule layers followed by tri-level dense convolution layer and an output layer. For experimentation, datasets are collected from more than 100 users for creation of training data samples of about 7769 comprising of 49 classes. Test samples of all the 49 classes are again collected separately from 3 to 5 users creating a total of 245 samples for novel patterns. It is inferred from performance evaluation; a loss of 0.66% is obtained in the classification process and for 43 classes precision of 100% is achieved with an accuracy of 99%. An average accuracy of 95% is achieved for all remaining 6 classes with an average precision of 89%.</p>

show abstract

Handwritten Character Recognition Using KNN and SVM Based Classifier over Feature Vector from Autoencoder

Cited by 9 publications

References 18 publications

Handwritten OCR for Indic Scripts: A Comprehensive Overview of Machine Learning and Deep Learning Techniques

Handwritten OCR for Indic Scripts: A Comprehensive Overview of Machine Learning and Deep Learning Techniques

Meta‐feature based few‐shot Siamese learning for Urdu optical character recognition

Robust recognition technique for handwritten Kannada character recognition using capsule networks

Contact Info

Product

Resources

About