流形核与LPP相结合的毛杆折痕识别方法  

Feather Quill Crease Recognition Method by Combing Manifold Kernel with LPP

作　　者：岳洪伟[1,2] 汪仁煌[2] 金迎迎[1] 明俊峰[2] 

机构地区：[1]五邑大学信息工程学院,广东江门529020 [2]广东工业大学自动化学院,广州510006

出　　处：《光电工程》2014年第2期47-52,共6页Opto-Electronic Engineering

基　　金：广东省自然科学基金(S2013040014993);广东省自然科学基金(S2012010010652);广东省科技计划项目(2012B020314005)

摘　　要：针对毛杆折痕难以检测问题,将非线性流形的思想引入到折痕识别领域。提出运用流形核函数与局部保持投影相结合的方法进行毛杆特征提取。首先基于区域图像构造协方差矩阵作为图像特征,利用仿射不变度量作为样本点的距离测度。然后通过定义的黎曼核函数选择流形上的近邻点,使得近邻点的选择符合数据呈非线性流形的假设,并结合数据类别信息构造相应的核矩阵。最后利用局部保持投影算法对毛杆图像进行降维。实验结果表明,本文算法能够有效克服光照不均和残余绒毛等外部因素影响,具有较好的稳健性和较高的识别率。Aimed at the detection difficult problem of feather quill crease, the idea of non-linear manifold is introduced into crease target recognition. A feather quill crease recognition method based on locality preserving projection and manifold kernel function is proposed for feature extraction. Firstly, covariance matrices are computed as the crease descriptors of feather quill, and an affine invariance metric which is adopted to make the space meet the requirement of Riemannian manifold is used to measure the distance between the two samples. Secondly, the neighbors of a selected point can be determined by the proposed manifold kernel function to make choice of the nearest neighboring points in line with the hypothesis of data distribution with non-linear manifold. The kernel matrix is defined based on the manifold distance and category labels. Finally, the locality preserving projections algorithm is used to reduce the dimensionality of the feather quill images. The simulated experiment results suggest that the proposed algorithm is robust to the variation of illumination and residual noises image segmentation, and achieves better performance compared with many popular recognition algorithms.

关 键 词：羽毛杆折痕 黎曼流形 流形核 局部保持投影 

分 类 号：TL361[核科学技术—核技术及应用]