Abstract-- In this paper, we present a completely distributed algorithm for partitioning a given set of mobile nodes into clusters. The proposed algorithm tries to reduce the amoun...
In recent years the Hessian matrix and its eigenvalues became important in pattern recognition. Several algorithms based on the information they provide have been introduced. We r...
This discussion presents a new perspective of subspace independent component analysis (ICA). The notion of a function of cumulants (kurtosis) is generalized to vector kurtosis. Th...
Linear Discriminant Analysis (LDA) has been a popular method for extracting features that preserves class separability. The projection functions of LDA are commonly obtained by max...
The problem of the independence and completeness of rotation moment invariants is addressed in this paper. First, a general method for constructing invariants of arbitrary orders ...