Odia character recognition system: A study on feature extraction and classification techniques

Singh, Sunny; Sarangi, Pradeepta Kumar; Singla, Chaitanya; Sahoo, Ashok Kumar

doi:10.1016/j.matpr.2020.04.680

Cited by 12 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The detection rate of an image is positively correlated with its picture quality, thus images with higher picture quality exhibit a greater detection rate compared to unprocessed noisy images. Consequently, the process of extracting features from unprocessed images poses challenges due to its impact on the efficiency of OCR [22].…”

Section: Roi Extractionmentioning

confidence: 99%

A Deep Learning Based Offline Optical Character Recognition Model for Printed Ottoman Turkish

AL-KHAFFAF,

ATİLA

2023

Technium

View full text Add to dashboard Cite

Developing efficient optical character recognition (OCR) systems for printed Ottoman text is a problem since current OCR models created for Arabic have restrictions that make it difficult to be performed. The performance of these models has been shown to be low when used for the recognition of Ottoman text. It has also been shown that these models that have been subjected to specialized training on Ottoman text have produced results that are not sufficient. In this study, an analysis of printed Ottoman Turkish documents in the Matbu font is conducted using a deep learning model that is proposed. Through the use of an end-to-end trainable architecture that integrates convolutional neural networks (CNNs) with bidirectional long short-term memory (BiLSTM) units, this study proposes an efficient solution to the Ottoman optical character recognition (OCR) issue. Experimental results show that the proposed model achieved overall scores for accuracy, sensitivity, and precision of 99.6%, 87.1%, and 93.3% on the test dataset respectively.

show abstract

Section: Roi Extractionmentioning

confidence: 99%

A Deep Learning Based Offline Optical Character Recognition Model for Printed Ottoman Turkish

AL-KHAFFAF,

ATİLA

2023

Technium

View full text Add to dashboard Cite

show abstract

“…In this step, records from feature extraction step along with class label of each record are given to the classifier for training of the classifier. [ 16–19 ] Then, testing records are provided to the classifier for predication. The prediction results are obtained in the form of confusion matrix and then various metrics are calculated to evaluate the classifier(s) used in the experiments.…”

Section: Classification Processmentioning

confidence: 99%

Indian Sign Language Recognition Using Ensemble Based Classifier Combination

Sahoo

Sarangi

Yadav

2022

Macromolecular Symposia

Self Cite

View full text Add to dashboard Cite

Indian Sign Language (ISL) is an alternate to written and direct communication language used in India and Indian subcontinent. It is commonly used by deaf and mute people who cannot speak and hear. The ISL is a new sign language as compared to other sign languages used in developed countries. There is a need for automatic recognition of any sign language including ISL due to its application aspects in current times. The automation of ISL will helpful to both communities (those who can interact only through ISL and those who do not know ISL at all), because in many places the unblessed people facing difficulty in communicating like at airports, railway stations, banks, hospitals, and other places. In this paper, the authors have tried to develop an automatic conversion from ISL signs to text in the domain of digits. The system requires a digital camera to capture the ISL signs and the underlying software will interpret the meaning of the image. To develop such a system, pattern recognition concept including image processing techniques have been used. Important features from the ISL signs are extracted using Histogram Oriented Gradient (HOG) technique. To classify the input images, two classification techniques have been used in the experiments namely Tree Bagger and Fit Ensemble.

show abstract

“…[8] has reviewed character recognition techniques for different languages in India and he observed that feature extraction technique is the most important step for character recognition. S. Singh et al [17] have surveyed extensively about the existing works in the field of Odia character recognition and observed that researchers have used features likeZernike moments [18], genetic algorithm [19], DCT and DWT [20][21], Standard deviation and zone centroid average distance-based feature matrix [22], a feature extracted using LU factorization [23][24], geometric features like centroid, shadow-based features and distance-based features [25], PCA [26] and rectangular HOG [27]. Through various research reviews it has been observed that HOG with SVM have been successful with character recognition on a larger dataset.…”

Section: Related Workmentioning

confidence: 99%

Feature Selection for Character Recognition of Handwritten Devanagari and Odia Scripts

Jha¹,

Parvathi²

2020

International Journal of Engineering Research and Technology

View full text Add to dashboard Cite

Handwritten character recognition has been a challenging task and it finds uses in many real time scenarios such as language translation, automatic Braille transliteration, automatic cheque book reading, automatic scanning of handwritten forms etc. This paper takes two Indian scripts namely Devanagari and Odia for handwritten character recognition and compares different gradient based features such as LBP, LDP, HOG and LOOP which can be used for learning the character pattern by a support vector classifier on various parameters. The paper uses two existing databases of handwritten Devanagari and Odia characters to train the support vector classifiers and compares the results of various features selection. It proposes the best possible feature for Devanagari and Odia character recognition based on graphical comparisons of parameters such as accuracy, training time and recognition rate. The maximum accuracy achieved on the Devanagari dataset of 92000 characters is 95.65% and the maximum accuracy achieved on the relatively small Odia dataset of 15400 characters is approximately 99%. The paper further investigates for the Devanagari characters getting misclassified more frequently.

show abstract

Odia character recognition system: A study on feature extraction and classification techniques

Cited by 12 publications

References 15 publications

A Deep Learning Based Offline Optical Character Recognition Model for Printed Ottoman Turkish

A Deep Learning Based Offline Optical Character Recognition Model for Printed Ottoman Turkish

Indian Sign Language Recognition Using Ensemble Based Classifier Combination

Feature Selection for Character Recognition of Handwritten Devanagari and Odia Scripts

Contact Info

Product

Resources

About