Vein matching method based on logarithm polar coordinate transformation

Abstract

The invention relates to a vein matching method based on logarithm polar coordinate transformation. According to the method, an image is pre-processed to obtain a vein line picture and a single pixel line picture, matching alternate regions of a template image and a sample image are designated according to finger vein line picture structure characteristics, then similarity measure of an LPT transformation result of the matching alternate regions of the template image and the sample image is calculated, and ultimately a matching result is judged according to the similarity measure and relative position distribution between the regions.

Description

technical field [0001] The present invention mainly relates to a vein matching method based on logarithmic polar transformation (LPT, Log-Polar Transform). Background technique [0002] The method of personal identity verification based on biometric identification is a research hotspot in the field of security protection. The vein feature is a biological feature, hidden inside the human body, unique and difficult to replicate, and is expected to become a new generation of universal personal identity verification features. Finger vein recognition mainly adopts global feature matching method, local feature matching method and multi-feature fusion matching method. Common among global feature matching methods are template matching methods and spatial projection methods (PCA, LDA, etc.). Template matching is generally based on vein binary images, which are robust to translation and plane rotation, but are not suitable for affine transformation caused by spatial rotation; in add...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the recognition results of existing vein recognition methods are easily affected by changes in finger placement and individual differences in imaging equipment, a vein matching method based on LPT is proposed.

Images