A Report on Uncorrelated Multi linear Principal Component Analysis Plus Extreme Learning Machine to Deal With Tensorial Data

作者：Zhang, Fan^*; Fan, Yao-Ling; Xu, Li

来源：Journal of Computational and Theoretical Nanoscience, 2015, 12(7): 1258-1262.

DOI：10.1166/jctn.2015.3882

• Pattern Recognition
• Uncorrelated Multi linear Principal Component Analysis
• Extreme Learning Machine
• Classification
• Feature Extraction

Subspace learning is an important direction in computer vision research. In this paper, a new method of tensor objects recognition based on uncorrelated multilinear principal component analysis (UMPCA) and extreme learning machine (ELM) is proposed. Because of mostly input data sets for pattern recognition are naturally multi-dimensional objects, UMPCA seeks a tensor-to-vector projection that captures most of the variation in the original tensorial input while producing uncorrelated features through successive variance maximization. A subset of features is extracted and the classifier ELM with extremely fast learning speed is then applied to achieve better performance. Extensive experiments are performed using challenging database and results are compared against state-of-the-art techniques.

• 出版日期2015-7