A System for Off-Line Oriya Handwritten Character Recognition Using Curvature Feature

Jayasingh, Bipin Bihari; Patra, Manas Ranjan

doi:10.1109/icit.2007.63

Cited by 51 publications

(21 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same Oriya data as used in this paper was segmented by connected components and used to train a CRF obtaining 93% character error rate on the same testset [31]. Chaudhuri and Pal reported a character recognition accuracy of 96.3% on printed Oriya dataset of basic characters [32] using a segmentation based, script dependent approach. The dataset used by us is full word sequenes from scanned document images from digitized old books which is highly complicated as compared to their dataset.…”

Section: Resultsmentioning

confidence: 99%

Text recognition using deep BLSTM networks

Ray

Rajeswar

Chaudhury

2015

2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)

View full text Add to dashboard Cite

This paper presents a Deep Bidirectional Long Short Term Memory (LSTM) based Recurrent Neural Network architecture for text recognition. This architecture uses Connectionist Temporal Classification (CTC) for training to learn the labels of an unsegmented sequence with unknown alignment. This work is motivated by the results of Deep Neural Networks for isolated numeral recognition and improved speech recognition using Deep BLSTM based approaches. Deep BLSTM architecture is chosen due to its ability to access long range context, learn sequence alignment and work without the need of segmented data. Due to the use of CTC and forward backward algorithms for alignment of output labels, there are no unicode re-ordering issues, thus no need of lexicon or postprocessing schemes. This is a script independent and segmentation free approach. This system has been implemented for the recognition of unsegmented words of printed Oriya text. This system achieves 4.18% character level error and 12.11% word error rate on printed Oriya text.

show abstract

Section: Resultsmentioning

confidence: 99%

Text recognition using deep BLSTM networks

Ray

Rajeswar

Chaudhury

2015

2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)

View full text Add to dashboard Cite

show abstract

“…Neural network is found very computationally expensive for a character recognition purpose [12]. Pal et al [14] have presented an offline handwritten Oriya character recognition system using curvature features. Bhattacharya et al [3] have presented an online handwritten Bangla character recognition system.…”

Section: Related Workmentioning

confidence: 99%

A Novel Hierarchical Technique for Offline Handwritten Gurmukhi Character Recognition

Kumar

Jindal

Sharma

2014

Natl. Acad. Sci. Lett.

View full text Add to dashboard Cite

The increasing need of a handwritten character recognition system in the Indian offices such as banks, post offices and so forth, has made it an imperative field of research. In present paper, Authors have presented a novel hierarchical technique for isolated offline handwritten Gurmukhi character recognition. A robust feature set of 105 feature elements is proposed under this work for recognition of offline handwritten Gurmukhi characters using four types of topological features, namely, horizontally peak extent features, vertically peak extent features, diagonal features, and centroid features. For classification Support Vector Machines (SVMs) classifier has been used in this work. SVMs classifier has been considered with four different kernels, namely, linear kernel, polynomial kernel, RBF kernel and sigmoid kernel. For training and testing of a classifier, we have used 3,500 samples of isolated offline handwritten Gurmukhi characters written by one hundred different writers. Maximum recognition accuracy of 91.80 % have been achieved with proposed technique, while using PCA feature set and SVM with a linear kernel classifier.

show abstract

“…Then, based on the feature vector, characters are classified into various groups using different techniques such as mean distance measure, artificial neural network, support vector machine, etc. Thereafter, in the recognition phase, again feature vector of the test characters is generated and needs to match [8,9].…”

Section: Recognition Techniquesmentioning

confidence: 99%

Odia Running Text Recognition Using Moment-Based Feature Extraction and Mean Distance Classification Technique

Nayak

2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Optical character recognition (OCR) is a process of automatic recognition of character from optically scanned documents for the purpose of editing, indexing, searching, as well as reduction in storage space. Development of OCR for an Indian script is an active area of research today because the presence of a large number of letters in the alphabet set, their sophisticated combinations, and the complicated grapheme's they formed is a great challenge to an OCR designer. We are trying to develop the OCR system for Odia language, which is used as official language of Odisha (formerly known as Orissa). In this paper, we attempt to recognize the vowels, consonants, matras, and compound characters of running Odia script. At first, the given scanned text is segmented into individual Odia symbols, then, extract corresponding feature vectors, using two-dimensional moments and Hough transform (based on topological and geometrical properties), which are used to classify and recognize the symbol. We found that the proposed model can recognize up to 100 % running test having no touched characters.

show abstract

A System for Off-Line Oriya Handwritten Character Recognition Using Curvature Feature

Cited by 51 publications

References 9 publications

Text recognition using deep BLSTM networks

Text recognition using deep BLSTM networks

A Novel Hierarchical Technique for Offline Handwritten Gurmukhi Character Recognition

Odia Running Text Recognition Using Moment-Based Feature Extraction and Mean Distance Classification Technique

Contact Info

Product

Resources

About