A KNN Improved Art Network Approach for Handwritten Character Recognition under Noise

One of the major application of Image processing is the character recognition. Character recognition is effective to convert the image character to text. This recognition process is more challenging in case of noisy or disturb image. In this work, a hybrid approach is suggested to perform this recognition effectively in case of disturbed noise image. The work is divided in three main stages. In first stage, the image improvement is performed by using denoising algorithm. In second stage, the image feature extraction is done to identify the character ROI and the feature points. This feature extraction is defined using KNN approach. At the final stage, Art network is applied to perform the image classification and recognition under effective vigilance ratio. The obtained results from system show the effective recognition rate.