为了更高效的处理高维数、高复杂性的非线性数据,发现其嵌入在源数据空间中的本维特征,提出了基于局部光滑逼近思想的流形学习算法,通过局部线性误差逼近最小化,实现将高维数据映射到低维空间.在FREY人脸数据库上进行降维实验,证明了该方法的可行性和有效性.
In order to process the nonlinear data with high-dimensional and high complexity efficiently,and found the dimensional characteristic embedded in the source data space,the manifold learning algorithm based on Locally smooth approximating was proposed on the basis of the analysis of the typical manifold learning algorithm.The dimensionality reduction of the high-dimensional nonlinear data is achieved by locally linear error approximating minimum.Through the experiments on dimensionality reduction in the FREY face database,the results show that the feasibility and effectiveness of manifold learning method applying to face image data processing.