摘要
在指纹数据库规模不断增大的情况下,指纹分类对于提高指纹识别的速度和准确率十分重要.本文提出一种利用指纹纹理信息的方法将指纹分为五大类.该方法利用指纹的中心点将指纹模式区分为四个部分并求取每一部分的局部二值模式方差,然后利用K近邻分类器进行分类.实验证明该方法具有良好的分类性能.
In the scale of fingerprint database is increasing, fingerprint classification is crucial for improving the speed and accuracy of fingerprint recognition.This paper presents a method based on the fingerprint texture ,and divides the fingerprint image into five categories.According to the core points,a fingerprint image is divided into four regions,each region is modeled with the distribution of the local binary pattern variance （LBPV） values ,and then it is classified by KNN classifies.Experiments verify that the method has good classification performance.
出处
《南华大学学报（自然科学版）》
2017年第1期57-62,共6页
Journal of University of South China：Science and Technology
关键词
局部二值模式(LBP)
K近邻
指纹分类
中心点
local binary pattern
k-nearest neighbor
fingerprint classification
core point
