ANN Implementation for Classification of Noisy Numeral Corrupted By Salt and Pepper Noise
Smita K. Chaudhari, Girish Kulkarni · International journal of advanced research in computer science and electronics engineering · 2012
Neural Network (NN) is information processing paradigm that is inspired by the way biological nervous systems, such as the brain, process information. Neural Networks are known to be capable of providing good recognition rate in presence of noise. Neural Network with various architectures and Training algorithms have successfully been applied for letter or character recognition [1]. Numerals Recognition is one of the artificial intelligence applications which provide an important fundamental for various advanced applications, including information retrieval and human-computer interaction applications. The neural networks are also able to extract meaningful features of the digits, such as edges. Handwritten recognition is complex due to large variation of handwritten style whereas printed character recognition is also difficult due to increase number fonts. This paper uses hamming netwok to recognize noisy numerals. The proposed algorithm will design a system which associates every fundamental pattern with itself. That is, when presented with x i as input, the system should produce x i at the output. In addition, when presented with a noisy (corrupted) version of x i at the input, the system should also produce x i at the output. The recognition results of the noisy numeral showed that the network could recognize normal numerals with 100% accuracy, numerals added with salt and pepper noise at average of 89%.