Character recognition for Writing Robot Control using ANFIS

Abstract

Robotics has dominates the industrial world since industrial revolution, due to its qualities in precision and accuracy. This paper is designed to display these qualities in a form of a writing robot. Image processing, character recognition, path planning and theta deduction are studied in this project. This paper served as discussion in image processing techniques used for writing robot and neural network to classify the letters. Letters are restricted to uppercase, and in a form of image. Image is converted into binary, which then letters are separated to form an image matrix. Image matrix will serve as training data for neural network. Performances of neural network are evaluated using test set prepared, to determine the scope of font recognizable using neural network.