In this paper, a novel algorithm for feature extraction, named supervised kernel locally principle component analysis (SKLPCA), is proposed. The SKLPCA is a non-linear and supervised subspace learning method, which maps the data into a potentially much higher dimension feature space by kernel trick and preserves the geometric structure of data according to prior class-label information. SKLPCA ...

We study the space of all compact structures on a two-dimensional square lattice of size N = 6 × 6. Each structure is mapped onto a vector in N dimensions according to a hydrophobic model. Previous work has shown that the designabilities of structures are closely related to the distribution of the structure vectors in the N -dimensional space, with highly designable structures predominantly fou...

It is well known that solutions to the nonlinear independent component analysis (ICA) problem are highly non-unique. In this paper we propose the “minimal nonlinear distortion” (MND) principle for tackling the ill-posedness of nonlinear ICA problems. MND prefers the nonlinear ICA solution with the estimated mixing procedure as close as possible to linear, among all possible solutions. It also h...

The traditional NCC algorithm has considerable computational burden, but is with high rate of matching accuracy when people use it in image matching. The PCA algorithm has high matching speed and can be anti-interference when it is used in image matching. Based on experiment, we propose one improved algorithm which can greatly reduce the matching time as well as get high rate of matching accuracy.