Review on Offline Handwritten Character Recognition using Feed Forward Neural Network and SURF Feature

Abstract: Image pre-processing is the name for operations on images at the lowest level of abstraction whose aim is an improvement of the image data that suppress undesired distortions or enhances some image features important for further processing. It does not increase image information content. Image pre-processing tool, created in Matlab, realizes many brightness transformations and local pre-processing methods. The proposed solutions focus on applying Neural Network Algorithm model for character recognition. The primary function of which is to retrieve in a character stored in memory, when an incomplete or noisy version of that character is presented. The idea is to create a theoretical and practical basis of preprocessing for character recognition using forward-feed neural networks. The Feed Forward Algorithm gives insight into the enter workings of a neural network; followed by the Back Propagation Algorithm which compromises Training and Testing.

Keywords: Character Recognition, Feed Forward Neural Network, Multi layer Perceptron, Supervised learning, SURF Feature.