Non-Correlated Character Recognition using Artificial Neural Network
This work addresses offline recognition of handwritten English characters without linear relationships, but it is incremental as it applies standard ANN methods to a specific dataset.
The paper tackled handwritten English character recognition for non-correlated characters using an artificial neural network, achieving a maximum recognition rate of 85%.
This paper investigates a method of Handwritten English Character Recognition using Artificial Neural Network (ANN). This work has been done in offline Environment for non correlated characters, which do not possess any linear relationships among them. We test that whether the particular tested character belongs to a cluster or not. The implementation is carried out in Matlab environment and successfully tested. Fifty-two sets of English alphabets are used to train the ANN and test the network. The algorithms are tested with 26 capital letters and 26 small letters. The testing result showed that the proposed ANN based algorithm showed a maximum recognition rate of 85%.