An effective method for number and letter character recognition

Abstract

An effective algorithm for number and letter character recognition is proposed in this paper. Our algorithm employs template matching, but it unlike traditional template matching method using the original pixel value to match. Our algorithm draws some features from the original image, and then obtains an eigenvector of 192 dimensions. Before drawing features, the image is disposed using math morphologic algorithm. And then measures the Euclidean distance between the sample vector and the template vector. Then we can obtain the result of recognition. A larger number of experiments prove that this algorithm own high performance and robustness. It can recognize characters which have high similarities, for example, 8 , B , R ,O and Q. This algorithm also tolerates the slightly tilt of the image. The recognition rate of this algorithm is 99.25%.