An orthogonal regularized CCA learning algorithm for feature fusion

作者：Hou Shudong^*; Sun Quansen

来源：Journal of Visual Communication and Image Representation, 2014, 25(5): 785-792.

DOI：10.1016/j.jvcir.2014.01.009

• Canonical correlation analysis
• Orthonormalization
• Regularization
• Feature fusion
• Partial least squares
• Dimensionality reduction
• Feature extraction
• Pattern recognition

Canonical correlation analysis (CCA) aims at extracting statistically uncorrelated features via conjugate orthonormalization constraints of the projection directions. However, the formulated directions under conjugate orthonormalization are not reliable when the training samples are few and the covariance matrix has not been exactly estimated. Additionally, this widely pursued property is focused on data representation rather than task discrimination. It is not suitable for classification problems when the samples that belong to different classes do not share the same distribution type. In this paper, an orthogonal regularized CCA (ORCCA) is proposed to avoid the above questions and extract more discriminative features via orthogonal constraints and regularized parameters. Experimental results on both handwritten numerals and face databases demonstrate that our proposed method significantly improves the recognition performance.

• 出版日期2014-7
• 单位中国电子科技集团公司第三十八研究所; 南京理工大学