Classification of handwritten signatures based on boundary tracing

Abstract

The paper presents a system for offline classification of handwritten signatures. The algorithm is based on boundary tracing technique for extracting characteristic features. Outer and inner boundaries are treated separately. The upper and lower parts of the boundaries are extracted to form two sequences ofpoints. Three algorithms fo r calculating feature vectors are applied based on у coordinate, distances between consecutive points and from polar coordinates system. Experiments on classification of the resulted vectors were carried outby means of Dynamic Time Warping algorithm using window and slope constraints.