International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 119 - Number 24 |
Year of Publication: 2015 |
Authors: Tarandeep Kaur, Amit Chabbra |
10.5120/21385-4391 |
Tarandeep Kaur, Amit Chabbra . Genetic Algorithm Optimized Neural Network for Handwritten Character Recognition. International Journal of Computer Applications. 119, 24 ( June 2015), 22-26. DOI=10.5120/21385-4391
Handwritten Character Recognition is well known problem which has many real world applications. Many solutions have already been proposed using various techniques (neural networks, fuzzy rules etc. ) over a period of time, but no one is able to achieve 100 percent accuracy rate. Involvement of various organizations for research on handwriting recognition has been significantly exaggerated over last few decades. Solution is required which can provide higher accuracy rate in lesser amount of computation time. This paper covers introduction to problem and various terms used, proposed solution based upon Neural Networks whose weights have been optimized using Genetic Algorithm (GA) with newly designed fitness function and performance comparison of proposed design with existing techniques various constraints.