Color text image binarization based on binary texture analysis

In this paper, a novel binarization algorithm for color text images is presented. This algorithm effectively integrates color clustering and binary texture analysis, and is capable of handling situations with complex backgrounds. In this algorithm, dimensionality reduction and graph theoretical clustering are first employed. As a result, binary images related to clusters can be obtained. Binary texture analysis is then performed on each candidate binary image. Two kinds of effective texture features, run-length histogram and spatial-size distribution related, respectively, are extracted and explored. Cooperating with a linear discriminant analysis classifier, the optimal candidate for the best binarization effect is obtained. Experiments with images collected from the Internet have been carried out and compared with existing techniques. Both show the effectiveness of the algorithm.