An Improved Characters Recognition Approach Using Fast Determinant Recursion

Abstract

The recognition technology of characters is one of the important technologies of intelligent transportation. At present, one of the problems that the technology needs to solve is that the recognition speed needs to be improved. The fast determinant recursion method is originally used for adaptive and fast processing of radar signals. In this paper, FDR (Fast Determinant Recursion) is used to speed up identifying characters. The advantage of this method is that it does not need to construct the estimation of the covariance matrix of the sample, and does not need to inverse the matrix. In the case of ensuring the constant accuracy of the recognition, the algorithm can improve the recognition speed of the characters. Some measured data are used to verify the effectiveness of the method.