On the Benefits of Convolutional Neural Network Combinations in Offline Handwriting Recognition

2016 ◽  
Author(s):  
Dewi Suryani ◽  
Patrick Doetsch ◽  
Hermann Ney
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Handwriting Recognition ◽  
Offline Handwriting Recognition

A FULLY AUTOMATED OFFLINE HANDWRITING RECOGNITION SYSTEM INCORPORATING RULE BASED NEURAL NETWORK VALIDATED SEGMENTATION AND HYBRID NEURAL NETWORK CLASSIFIER

2004 ◽  
Vol 18 (07) ◽  
pp. 1267-1283 ◽  
Author(s):  
MOUMITA GHOSH ◽  
RANADHIR GHOSH ◽  
BRIJESH VERMA
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Solution Method ◽  
Handwriting Recognition ◽  
Recognition System ◽  
Neural Network Classifier ◽  
Hybrid Neural Network ◽  
Rule Based ◽  
Extraction Algorithm ◽  
Offline Handwriting Recognition

In this paper we propose a fully automated offline handwriting recognition system that incorporates rule based segmentation, contour based feature extraction, neural network validation, a hybrid neural network classifier and a hamming neural network lexicon. The work is based on our earlier promising results in this area using heuristic segmentation and contour based feature extraction. The segmentation is done using many heuristic based set of rules in an iterative manner and finally followed by a neural network validation system. The extraction of feature is performed using both contour and structure based feature extraction algorithm. The classification is performed by a hybrid neural network that incorporates a hybrid combination of evolutionary algorithm and matrix based solution method. Finally a hamming neural network is used as a lexicon. A benchmark dataset from CEDAR has been used for training and testing.

scholarly journals Arabic Handwritten Digit Recognition using Convolutional Neural Network

2020 ◽  
Vol 8 (6) ◽  
pp. 1187-1190
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Handwriting Recognition ◽  
Data Entry ◽  
Arabic Numeral ◽  
Experimental Result ◽  
Digit Recognition ◽  
Average Accuracy ◽  
Handwritten Digit ◽  
Arab League

Arabic is the most widely used language in the world, especially the Arab League Country. Of course, in those countries often use Arabic numeral in banks and business applications, postal zip code and data entry application. This research has focused on handwriting recognition of Arabic numeral that has unlimited variation in human handwriting such as style and shape. The proposed method on the deep learning technique is Convolutional Neural Network. LeNet-5 architect also used in training and recognizing the handwritten image of Arabic numeral as much as 70000 images derived from MADbase dataset. The experimental result on 10000 images of database used is by comparing the number of epoch in training process yields, and the average accuracy is 97.67%.

Sign in / Sign up

Export Citation Format

Share Document