Effective multiple-features extraction for off-line SVM-based handwritten numeral recognition

In this paper, a multiple features extraction technique for the recognition of handwritten numbers is proposed. The proposed technique mainly extracts direction information from the structure of contours of each handwritten number and the direction information is integrated with a technique for detecting transitions among pixels and counting the number of cross lines in the lined image of offline handwritten numbers. The combinational technique used in the recognition with a Support Vector Machine (SVM) [13] classifier provides recognition rates up to 98.99%. This proposed technique also uses SVM for determining the effective features extracted from the multiple features extraction of the handwritten number recognition.