Classification of German scripts by adjacent local binary pattern analysis of the coded text

The paper proposes a script classification method which is based on textural analysis of the script types. In the first stage, each letter is coded by the equivalent script type, which is defined by its baseline position. Obtained coded text is subjected to the adjacent local binary pattern analysis to extract the features. The result shows the diversity of the extracted features between scripts, which makes the feature classification easier. It is the basis for decision-making process of the script identification by automatic classification. The proposed method is tested on an example of synthetic and historical German printed documents written in Antiqua and Fraktur scripts. The experiment shows very positive results, which proved the correctness of the proposed algorithm.