Palmprint Matching Using Line Features

Palmprint-based personal identification is regarded as an effective method for automatically recognizing a person´s identity. It is well-known for several advantages such as stable line features, low-resolution imaging, low-cost capturing device, and user friendliness. This paper presents a new approach to identify palmprints using modified line-based Hausdorff distance. Line edge map (LEM) of palmprint is extracted as the feature used for identification. Length, position, and orientation information of the palm lines are fully utilized for palmprint matching. In contrast to the existing methods, our approach achieves higher accuracy. It is able to yield 100% identification accuracy and 0% EER using the public palmprint database provided by PolyU