A.A. Ghaibeh, S. Kuroyanagi, and A. Iwata (Japan)
CombNET, Neural networks, Character recognition
The large-scale neural network CombNET-II has been used in many applications and provided good results especially for problems where the number of classes is very large such as the recognition of Asian languages' printed and handwritten character recognition. However CombNET-II still requires a large amount of memory used for the training database and feature space. Here we propose a revised version of CombNET-II with a considerably lower memory requirement, this is very important for implying CombNET-II in small devices such as electronic dictionaries or diaries. Testing our proposed model using handwritten digits shows that the required memory might be reduced by almost 50% without significant decrease in the recognition rate.
Important Links:
Go Back
