scholarly journals Bengali Longhand Character Recognition using Fourier Transform and Euclidean Distance Metric

2018 ◽  
Vol 3 (7) ◽  
pp. 67
Author(s):  
Mousumi Hasan Mukti ◽  
Quazi Saad-Ul-Mosaher ◽  
Khalil Ahammad
Keyword(s):  
Pattern Recognition ◽  
Fourier Transform ◽  
Character Recognition ◽  
Euclidean Distance ◽  
Great Part ◽  
Research Area ◽  
Input Image ◽  
Handwritten Character ◽  
Input Dataset ◽  
Implementation Tool

Handwritten Character Recognition (HCR) is widely considered as a benchmark problem for pattern recognition and artificial intelligence. Text matching has become a popular research area in recent days as it plays a great part in pattern recognition. Different techniques for recognizing handwritten letters and digits for different languages have already been implemented throughout the world. This research aims at developing a system for recognizing Bengali handwritten characters i.e. letters and digits using Fourier Transform (FT) and Euclidean distance measurement technique. A dataset with 800 handwritten character texts from different people has been developed for this purpose and these character texts are converted to their equivalent printed version to implement this research. MATLAB has been used as an implementation tool for different preprocessing techniques like cropping, resizing, flood filling, thinning etc. Processed text images are used as input to the system and they are converted to FT. Handwritten character of different person may be of different style and angle. The input dataset is collected from various types of people including age level from 5 to 70 years, from different professions like pre-schooling students, graduate students, doctors, teachers and housewives. So, to match the input image with printed dataset (PDS) each printed data is rotated up to 450 left and right and then their FT is computed. The Euclidean distance among the input image and the rotated 30 images of each printed text are taken as intermediate distance set. The minimum value of Euclidean distance for a character is used to recognize the targeted character from the intermediate set. Wrongly detected texts are not thrown away from the system rather those are stored in the named character or digits file so that those can be used in future for deep learning. By following the proposed methodology, the research has achieved 98.88% recognition accuracy according to the input and PDS.

Lanna Handwritten Character Recognition on Historical Documents Using Feature Extraction

2012 ◽  
Vol 263-266 ◽  
pp. 2553-2560
Author(s):  
Krisda Khankasikam
Keyword(s):  
Feature Extraction ◽  
Character Recognition ◽  
Cross Validation ◽  
Research Area ◽  
Input Image ◽  
Historical Documents ◽  
Validation Data ◽  
Handwritten Character ◽  
Starting Point ◽  
Character Feature

The objective of this paper is to develop methodology that can recognize the Lanna handwritten character on historical documents by using character feature extraction technique. Historical documents are national treasures. Insignificant effort has been made to preserve Lanna historical documents. Other nations such as Egypt, China and Greece are investing a large effort in restoring and preserving their national historical documents. As a starting point, the focus is on using one Lanna historical document for performing experiments and verifying recognition methods available in this research area. The proposed system consists of three modules, which are image preprocessing module, feature extraction module and character recognition module. The details of each module are following: first, the input image is transformed into a suitable image for feature extraction module. Second, the proposed system extracts character features from the image. Finally, the extracted character information, which is kept in form of bit string, is calculated a similarity value for recognition result. The experiment was conducted on more than 4,000 Lanna handwritten characters by using 10-fold cross-validation classification method which is using 3,600 for training characters and 400 for testing character. The cross-validation process is repeated 10 times, with each of the 10 subsets used exactly once as the validation data. The precision of the proposed system is around 89.73 percent.

scholarly journals PENGENALAN AKSARA BALI DENGAN PENDEKATAN METODE DIRECTION FEATURE DAN AREA BINARY OBJECT FEATURE

2013 ◽  
Vol 9 (1) ◽  
Author(s):  
Ni Made Ari Pratiwi ◽  
Widi Hapsari ◽  
Theresia Herlina R.
Keyword(s):  
Pattern Recognition ◽  
Character Recognition ◽  
Euclidean Distance ◽  
Human Being ◽  
Object Feature ◽  
Object Area ◽  
Binary Object ◽  
The Right ◽  
Object Features ◽  
Small Stroke

The rise of technology has been contributing advances to science, and to human being in making jobs far much easier to do, including pattern recognition. This research focused on character recognition by developing a system capable to recognise images of printed Balinese traditional character, which has a distinct feature of having perceptually similar characters, where each others are often differentiated only by a small stroke or a curve. The system itself took several processes to recognise a character. First, the image containing Balinese characters is preprocessed. Afterwards, two object features are extracted from the image: Direction and Binary Object Area. Both features then tested for similarity using Euclidean distance with the same features already obtained from the control images. From 573 characters tested to the system, 559 are recognized as characters and 526 are correctly recognized as the right character, which yields an overall accuracy of 91.8%. Recognition results are dependent to character spacing condition.

An Efficient Handwritten Character Recognition Using Quantum Multilayer Neural Network (QMLNN) Architecture

2021 ◽  
pp. 435-446
Author(s):  
Debanjan Konar ◽  
Suman Kalyan Kar
Keyword(s):  
Neural Network ◽  
Character Recognition ◽  
Input Image ◽  
Handwritten Character Recognition ◽  
Handwritten Character ◽  
Fuzzy Input ◽  
Feed Forward Network ◽  
Network Output ◽  
Fuzziness Measure ◽  
Neighborhood Topology

This chapter proposes a quantum multi-layer neural network (QMLNN) architecture suitable for handwritten character recognition in real time, assisted by quantum backpropagation of errors calculated from the quantum-inspired fuzziness measure of network output states. It is composed of three second-order neighborhood-topology-based inter-connected layers of neurons represented by qubits known as input, hidden, and output layers. The QMLNN architecture is a feed forward network with standard quantum backpropagation algorithm for the adjustment of its weighted interconnection. QMLNN self-organizes the quantum fuzzy input image information by means of the quantum backpropagating errors at the intermediate and output layers of the architecture. The interconnection weights are described using rotation gates. After the network is stabilized, a quantum observation at the output layer destroys the superposition of quantum states in order to obtain true binary outputs.

scholarly journals Handwritten Character Recognition Using Multiresolution Technique and Euclidean Distance Metric

2012 ◽  
Vol 03 (02) ◽  
pp. 208-214 ◽  
Author(s):  
Dileep Kumar Patel ◽  
Tanmoy Som ◽  
Sushil Kumar Yadav ◽  
Manoj Kumar Singh
Keyword(s):  
Character Recognition ◽  
Euclidean Distance ◽  
Distance Metric ◽  
Handwritten Character Recognition ◽  
Handwritten Character

HANDWRITTEN CHARACTER RECOGNITION USING NONSYMMETRICAL PERCEPTUAL ZONING

2007 ◽  
Vol 21 (01) ◽  
pp. 135-155 ◽  
Author(s):  
CINTHIA O. A. FREITAS ◽  
LUIZ S. OLIVEIRA ◽  
FLÁVIO BORTOLOZZI ◽  
SIMONE B. K. AIRES
Keyword(s):  
Character Recognition ◽  
Handwriting Recognition ◽  
Input Image ◽  
Alternative Strategy ◽  
Handwritten Character Recognition ◽  
Handwritten Character ◽  
Recognition Systems ◽  
Confusion Matrices

In this paper we present an alternative strategy to define zoning for handwriting recognition, which is based on nonsymmetrical perceptual zoning. The idea is to extract some knowledge from the confusion matrices in order to make the zoning process less empirical. The feature set considered in this work is based on concavities/convexities deficiencies, which are obtained by labeling the background pixels of the input image. To better assess the nonsymmetrical zoning we carried out experiments using four different zonings strategies. Experiments show that the nonsymmetrical zoning could be considered as a tool to build more reliable handwriting recognition systems.

Thai handwritten character recognition by Euclidean distance

2010 ◽  
Author(s):  
Chomtip Pornpanomchai ◽  
Pattara Panyasrivarom ◽  
Nuttakit Pisitviroj ◽  
Piyaphume Prutkraiwat
Keyword(s):  
Character Recognition ◽  
Euclidean Distance ◽  
Handwritten Character Recognition ◽  
Handwritten Character

An Efficient Handwritten Character Recognition Using Quantum Multilayer Neural Network (QMLNN) Architecture

2018 ◽  
pp. 262-276 ◽  
Author(s):  
Debanjan Konar ◽  
Suman Kalyan Kar
Keyword(s):  
Neural Network ◽  
Character Recognition ◽  
Input Image ◽  
Handwritten Character Recognition ◽  
Handwritten Character ◽  
Fuzzy Input ◽  
Feed Forward Network ◽  
Network Output ◽  
Fuzziness Measure ◽  
Neighborhood Topology

This chapter proposes a quantum multi-layer neural network (QMLNN) architecture suitable for handwritten character recognition in real time, assisted by quantum backpropagation of errors calculated from the quantum-inspired fuzziness measure of network output states. It is composed of three second-order neighborhood-topology-based inter-connected layers of neurons represented by qubits known as input, hidden, and output layers. The QMLNN architecture is a feed forward network with standard quantum backpropagation algorithm for the adjustment of its weighted interconnection. QMLNN self-organizes the quantum fuzzy input image information by means of the quantum backpropagating errors at the intermediate and output layers of the architecture. The interconnection weights are described using rotation gates. After the network is stabilized, a quantum observation at the output layer destroys the superposition of quantum states in order to obtain true binary outputs.

scholarly journals Handwritten English Alphabet Recognition

2021 ◽  
Vol 9 (12) ◽  
pp. 2134-2139
Author(s):  
Poonam Bhanudas Abhale
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Literature Review ◽  
Character Recognition ◽  
Digital Form ◽  
Handwritten Character Recognition ◽  
Handwritten Character ◽  
English Alphabet ◽  
Recycling Systems ◽  
Segmentation Point

Abstract: Character recognition is a process by which a computer recognizes letters, figures, or symbols and turns them into a digital form that a computer can use. In moment’s terrain character recognition has gained a lot of attention in the field of pattern recognition. Handwritten character recognition is useful in cheque processing in banks, form recycling systems, and numerous further. Character recognition is one of the well- liked and grueling areas of exploration. In the unborn character recognition produce a paperless terrain. In this paper, we describe the detailed study of the being system for handwritten character recognition. We give a literature review on colorful ways used in offline English character recognition. Keywords: Character; Character recognition; Preprocessing; Segmentation; Point birth; Bracket; neural network; Convolution neural network.

scholarly journals Handwritten Gurmukhi Digit Prediction using Convolutional Neural Network with Keras and Tensorflow

2019 ◽  
pp. 777-784
Author(s):  
Sonia Flora ◽  
Divya Ebenezer Nathaniel
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Character Recognition ◽  
Research Area ◽  
Number Of Layers ◽  
Handwritten Character ◽  
Single Person ◽  
Intelligent Character Recognition

Intelligent Character Recognition is a term which is specifically used for the recognition of handwritten character or digit. It is a prominent research area of computer vision field of machine learning or deep learning which trained the machine to analyze the pattern of handwritten character image and identify it. Recognition of handwritten character is a hard process because single person can handwrite the same text in number of ways by making a little variation in holding the pen. Handwriting has no specific font style or size. It differs person to person or more specifically it differs how one is holding the pen. Deep Leaning has brought the breakthrough performance in this research area with its dedicated models like Convolutional Neural Network, Recurrent Neural Network etc. In this paper, we have trained model with Convolutional Neural Network with different number of layers and filters over 10,559 handwritten gurmukhi digit images and validate over 1320 images. Consequently we could achieve the maximum accuracy of 99.24%.

Sign in / Sign up

Export Citation Format

Share Document