Performance enhancement of the pattern recalling efficiency of Hopfield neural network using genetic algorithm for cursive handwritten character recognition
by Sonal Bansal; Rinku Dixit
International Journal of Applied Pattern Recognition (IJAPR), Vol. 3, No. 1, 2016

Abstract: In this paper, we are describing the implementation of recalling cursive handwritten characters extracted through a series of segmentation steps performed on a cursive handwritten sentences done using Hopfield neural network and genetic algorithm (GA). The feature extraction has been performed using segmentation algorithm. A novel algorithm for handwriting segmentation has been proposed to get isolated characters as patterns. These patterns are stored in Hopfield neural network using Hebbian learning rule and the recalling is done using Hopfield neural network and evolutionary algorithm. GA is an evolutionary algorithm and is used to enhance the pattern recalling efficiency of Hopfield networks. The observation made from the simulated results is that our segmentation algorithm segments the word image into characters. Later on, these characters will be fed into Hopfield neural network for the pattern recognition. The recall efficiency of the Hopfield neural network can be sufficiently enhanced when supported with GAs.

Online publication date: Thu, 09-Jun-2016

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Applied Pattern Recognition (IJAPR):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com