Off-Line Chinese Signature Verification Segmentation and Feature Extraction

One of the key tasks in off-line Chinese signature verification is how to acquire the segments and its features from the signature image. In this paper we present a new method to solve the problem for random forgeries and simple forgeries. After the signature being binaried, normalized and thinned, the signature image is segmented to some segments and some segment chains. Every segment is represented by a set of seven features, and every segment chain includes a segment chain includes a series of segments. Using the features of segments and considering the impact of the segment chains, a similarity of the signature is computed. A verification rate of 91% has been gained.