The development of optical character recognition approach on the basis of joint application of probabilistic neural network and wavelet transform

Abstract

There are a lot of approaches to solve the problem of optical character recognition. One of them is the approach based on neural networks. The authors have proposed and implemented an approach of analyzing and classification based on wavelet transformation for reducing the feature space and probabilistic neural network for recognition. The main advantage of probabilistic neural network is the probabilistic significance of its output neurons which can be used to improve the quality of recognition. Optimal values of parameters were selected for such approach. The results of numerical experiments were analyzed and the time-performance of this approach was assessed. The set of noisy character images was used to assess the proposed approach. This method showed 98 % of acceptable recognition.

Keywords

• optical character recognition
• probabilistic neural network
• probabilistic assessment
• wavelet transform
• wavelets