Techniques for informationhidinghave become increasingly more sophisticated and widespread. With high-resolution digital images as carriers, detecting hidden messages has become c...
In this paper, we propose a method to select a discriminative set of image processing operations for Linear Discriminant Analysis (LDA) as an application of the use of generating ...
Abstract--High-dimensional data are common in many domains, and dimensionality reduction is the key to cope with the curse-of-dimensionality. Linear discriminant analysis (LDA) is ...
While null space based linear discriminant analysis (NLDA) obtains a good discriminant performance, the ability easily suffers from an implicit assumption of Gaussian model with sa...