A Palmprint ROI Segmentation Algorithm for Eliminating Wrist Interference

Palmprint region of interest (ROI) segmentation is an important step in palmprint recognition. Positioning successful ROI has a greater impact on subsequent feature extraction and matching. At present, the palmprint segmentation method mainly has problems in that the positioning points are not easy to determine, and the poor positioning stability for different images of the same person. Therefore, based on the existing algorithm that based on regulate angle, a palmprint ROI segmentation algorithm that eliminating wrist interference is proposed, which improving the method of locating finger valley points and selecting the little finger side edge line. A method of wrist segmentation is designed to effectively avoid the influence of the wrist on the positioning of the feature points. The experimental results finally show that the proposed algorithm is more accurate for palmprint ROI segmentation and has better positioning stability. The ROI extraction success rate can reach 99.80%.