Palmprint Matching using Pairwise Relative Angle Histogram

Abstract

In this paper, we propose a novel approach to retrieve line-patterns from large databases in a rotation and translation invariant manner, at the same time, tackle broken line problem. Line segments are extracted from an image as primitives. Each local structure is represented by a set of pair-wise angle relationships, which are simple, invariant to translation and rotation, robust to end-point erosion, segment error, and sufficient for discrimination. Experiment showed encouraging results which also implicate that line segments could provide sufficient information for palmprint recognition.