摘要
在毛杆折痕识别中,光照不均和边缘绒毛等干扰因素在一定程度上影响识别的准确性.为尽可能排除这些影响折痕识别的因素,将图像稀疏表示思想引入到折痕检测,提出了黎曼流形上的核稀疏表示方法进行目标识别.首先采用协方差矩阵描述子来表征图像,并利用Bregman散度作为黎曼流形上的距离度量,进而构造了流形上的核函数.然后通过核方法把样本数据映射到再生核希尔伯特空间,结合稀疏表示获取高维特征空间的核稀疏表示系数.最后建立了字典学习的数学模型,结合凸理论提出了字典学习的有效算法.实验结果表明了所提方法的有效性.
In feather quill crease detection, the existences of uneven illumination and fluff usually influence the detection accuracy. Ai- ming at the detection difficult problem, the idea of image sparse representation is introduced into the field of crease recognition. A feather quill crease recognition method based on sparse representation on Riemannian manifold is proposed for target recognition. First- ly, covariance matrices are computed as the crease descriptors of feather quill, and Bregman divergence which is adopted to make the space meet the requirement of Riemannian manifold is used to measure the distance between the two samples, then kernel function on Riemannian manifold is constructed. Secondly, the sample datasets are mapped into the reproducing kernel Hilbert space by kernel method, and kernel sparse representation coefficient of high dimensional feature space is obtained based on sparse representation. Final- ly, a mathematical model of dictionary learning is constructed and an efficient algorithm is proposed for dictionary learning according to the convex theory. The simulated experimental results verify the effectiveness of the proposed method which achieves better per- formance compared with many popular recognition algorithms.
出处
《小型微型计算机系统》
CSCD
北大核心
2015年第4期886-890,共5页
Journal of Chinese Computer Systems
基金
广东省自然科学基金项目(S2013040014993)资助
广东省自然科学基金项目(S2012010010652)资助
广东省科技计划项目(2012B020314005)资助
作者简介
岳洪伟，男，1979年生，博士，讲师，研究方向为图像处理、模式识别。E-mail：yuehongwei420@163．com金迎迎，女，1982年生，博士研究生，讲师，研究方向为微分拓扑学．
