Effective Kernel-based Dimensionality Reduction for Wood Defects Recognition
Dimensionality reduction is important preprocessing step in high-dimensional data analysis without losing much intrinsic information.The problem of the kernel based semi-supervised nonlinear dimensionality reduction called KNDR is considered for wood defects recognition.In this setting,domain knowledge in form of pairs constraints are adopted to specify whether pairs of instances belong to same class or not.KNDR can project the data onto a set of ”useful” features and preserve the structure of labeled and unlabeled data as well as the constraints defined in the embedded spaces,under which the projections of the data can be effectively partitioned from each other.We demonstrate the practical usefulness and high scalability of KNDR for data visualization and wood defects recognition through extensive simulation experiments.Experimental results show KNDR can achieve similar or even higher performances than the typical PCA,KPCA and KFD methods.
Semi-supervised Learning Wood Defects Recognition Dimensionality Reduction (Dis-) similar Constraints Local Binary Pattern (LBP)
Zhang Zhao Ye Ning
School of Information Technology, Nanjing Forestry University, Nanjing, 210037, China School of Information Technology, Nanjing Forestry University, Nanjing, 210037, China;School of Comp
国内会议
山东泰安
英文
562-564
2009-08-15（万方平台首次上网日期，不代表论文的发表时间）