摘要
提出一种基于核函数稀疏编码的掌纹分类方法。将局部二值模式特征作为核函数稀疏编码分类器的输入,通过核函数,在高维特征空间建立字典及稀疏编码分类器,完成对未知掌纹的分类。在香港理工大学掌纹库上实验表明,新方法较传统方法分类的精度更高。
A palmprint classification method based on kernel sparse coding is presented. Local binary patterns features are as input of kernel sparse coding classifier. By kernel function, builds a dictionary and sparse coding classifier in the high dimensional feature space. The experimental results on palmprint data of Hongkong Polytech University show that new method has higher classification accuracy than the tradition methods.
出处
《自动化技术与应用》
2017年第12期98-101,共4页
Techniques of Automation and Applications
关键词
掌纹识别
稀疏编码
核函数
局部二值模式
palmprint recognition
sparse representation
kernel function
local binary patterns
