Devanagari Handwritten Marathi Compound Character Recognition System

2021 ◽  
Vol 6 (3) ◽  
Author(s):  
Chitra Bhole
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Pattern Recognition ◽  
Feature Extraction ◽  
Character Recognition ◽  
Wavelet Transformation ◽  
Recognition System ◽  
Computational Time ◽  
Deep Convolution Neural Network ◽  
Time Required

Handwritten character recognition a field of research in AI, computer vision, and pattern recognition. Devanagari handwritten Marathi compound character recognition is most tedious tasks because of its complexity as compared to other languages. As compound character is combination of two or more characters it becomes challenging task to recognize it. However, the researchers used various methods like Neural Network, SVM, KNN, Wavelet transformation to classify the features of compound Marathi characters and tried to give the accuracy in the recognition of it. But the problem of feature extraction, and time required is large. In this paper I am proposing the Offline handwritten Marathi compound character recognition using deep convolution neural network which reduces the computational time and increases the accuracy.

FEATURE EXTRACTION IN CHARACTER RECOGNITION WITH ASSOCIATIVE MEMORY CLASSIFIER

1996 ◽  
Vol 10 (04) ◽  
pp. 325-348 ◽  
Author(s):  
MING ZHANG ◽  
CHING Y. SUEN ◽  
TIEN D. BUI
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Feature Extraction ◽  
Associative Memory ◽  
Data Reduction ◽  
Character Recognition ◽  
Recognition System ◽  
Extraction Process ◽  
The Neural Network ◽  
Series Of Experiments

A pattern recognition system mainly contains two functional parts, i.e. feature extraction and pattern classification. The success of such a system depends on not only the effectiveness of each of them, but also their operation in concert. The feature extraction process in a traditional recognition system has two major tasks, namely, to extract deformation invariant signals and to reduce data. When a neural network is used as a pattern classifier, however, an alteration in these basic objectives is needed. In particular, the consideration of data reduction will be replaced by that of the suitability of feature vectors to the neural network. In this paper, feature extraction algorithms in character recognition have been designed based on these principles. The improvements made by these algorithms have been demonstrated in a series of experiments which justify such a change in the fundamental objectives of the feature extraction process when an associative memory classifier is used.

scholarly journals Handwritten English Character Recognition and translate English to Devnagari Words

2019 ◽  
pp. 142-151
Author(s):  
Shivali Parkhedkar ◽  
Shaveri Vairagade ◽  
Vishakha Sakharkar ◽  
Bharti Khurpe ◽  
Arpita Pikalmunde ◽  
...  
Keyword(s):  
Pattern Recognition ◽  
Feature Extraction ◽  
Character Recognition ◽  
Computational Time ◽  
Feature Vectors ◽  
Form Processing ◽  
Knn Classifier ◽  
Gabor Feature ◽  
Handwritten Document ◽  
The Individual

In our proposed work we will accept the challenges of recognizing the words and we will work to win the challenge. The handwritten document is scanned using a scanner. The image of the scanned document is processed victimization the program. Each character in the word is isolated. Then the individual isolated character is subjected to “Feature Extraction” by the Gabor Feature. Extracted features are passed through KNN classifier. Finally we get the Recognized word. Character recognition is a process by which computer recognizes handwritten characters and turns them into a format which a user can understand. Computer primarily based pattern recognition may be a method that involves many sub process. In today’s surroundings character recognition has gained ton of concentration with in the field of pattern recognition. Handwritten character recognition is beneficial in cheque process in banks, form processing systems and many more. Character recognition is one in all the favored and difficult space in analysis. In future, character recognition creates paperless environment. The novelty of this approach is to achieve better accuracy, reduced computational time for recognition of handwritten characters. The proposed method extracts the geometric features of the character contour. These features are based on the basic line types that forms the character skeleton. The system offers a feature vector as its output. The feature vectors so generated from a training set, were then used to train a pattern recognition engine based on Neural Networks so that the system can be benchmarked. The algorithm proposed concentrates on the same. It extracts totally different line varieties that forms a specific character. It conjointly also concentrates on the point options of constant. The feature extraction technique explained was tested using a Neural Network which was trained with the feature vectors obtained from the proposed method.

Recognition of Vehicles on Geometric Morphology

2011 ◽  
Vol 217-218 ◽  
pp. 27-32
Author(s):  
Guo Feng Qin ◽  
Yu Sun ◽  
Qi Yan Li
Keyword(s):  
Computer Vision ◽  
Pattern Recognition ◽  
Feature Extraction ◽  
Traffic Management ◽  
Template Matching ◽  
Recognition System ◽  
Automatic Recognition ◽  
Vehicular Traffic ◽  
Image Pretreatment ◽  
Geometric Morphology

Detection of vehicles plays an important role in the area of the modern intelligent traffic management. And the pattern recognition is a hot issue in the area of computer vision. This article introduces an Automobile Automatic Recognition System based on image. It begins with the structures of the system. Then detailed methods for implementation are discussed. This system take use of a camera to get traffic images, then after image pretreatment and segmentation, do the works of feature extraction, template matching and pattern recognition, to identify different models and get vehicular traffic statistics. Finally, the implementation of the system is introduced. The algorithms of recognized process were verified in this application case.

scholarly journals Neural Network based Handwritten Character Recognition system

2018 ◽  
Vol 7 (03) ◽  
pp. 23761-23768 ◽  
Author(s):  
Savitha Attigeri
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Character Recognition ◽  
Recognition System ◽  
Feed Forward Neural Network ◽  
Handwritten Character Recognition ◽  
Data Set ◽  
Research Areas ◽  
Handwritten Character ◽  
The Neural Network

Handwritten character recognition has been one of the active and challenging research areas in the field of image processing and pattern recognition. It has numerous applications which include, reading aid for blind, bank cheques and conversion of any hand written document into structural text form. In this paper an attempt is made to recognize handwritten characters for English alphabets without feature extraction using multilayer Feed Forward neural network. Each character data set contains 26 alphabets. Fifty different character data sets are used for training the neural network. The trained network is used for classification and recognition. In the proposed system, each character is resized into 30x20 pixels, which is directly subjected to training. That is, each resized character has 600 pixels and these pixels are taken as features for training the neural network. The results show that the proposed system yields good recognition rates which are comparable to that of feature extraction based schemes for handwritten character recognition

scholarly journals Probabilistic Neural Network based Approach for Handwritten Character Recognition

2010 ◽  
pp. 105-109
Author(s):  
V.N. Manjunath Aradhya ◽  
S. K. Niranjan ◽  
G. Hemantha Kumar
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Character Recognition ◽  
Probabilistic Neural Network ◽  
Principal Component ◽  
Recognition System ◽  
Indian Language ◽  
South Indian ◽  
Extraction Scheme ◽  
Standard Techniques

In this paper, recognition system for totally unconstrained handwritten characters for south Indian language of Kannada is proposed. The proposed feature extraction technique is based on Fourier Transform and well known Principal Component Analysis (PCA). The system trains the appropriate frequency band images followed by PCA feature extraction scheme. For subsequent classification technique, Probabilistic Neural Network (PNN) is used. The proposed system is tested on large database containing Kannada characters and also tested on standard COIL-20 object database and the results were found to be better compared to standard techniques.

Sign in / Sign up

Export Citation Format

Share Document