Image compression using principal component neural networks

Penalty function-based joint diagonalization approach for convolutive blind separation of nonstationary sources

“…Then the perturbation of (18) reads tr (32) where is a small scalar, and is the real part of a complex value. On the other hand (33) where , and can be expanded separately diag diag (34) Omitting the time and frequency index for simplicity and regarding (5) and (9), (34) becomes (35) and (36), shown at the bottom of the page, where , were applied in (36).…”

“…We may choose a penalty function with the form of (see [39] for a similar application) or Tr (see [36] for a similar application), where is a diagonal matrix containing the Lagrangian multipliers.…”

A hybrid randomized algorithm for image compression

“…The R, G, and B matrices contain image color components, the weights α, β, and γ were determined with regards to the possibilities of human perception [2]. There is a huge amount of algorithms [1,2,5,7] based on various principles leading to the image compression.…”

“…There is a huge amount of algorithms [1,2,5,7] based on various principles leading to the image compression. Algorithms based on the image color reduction are mostly lossy but their results are still acceptable for some applications.…”

“…Among various qualified statistical procedures, the principal component analysis (PCA) is the most classical and popular one. Since it uses only simple linear algebra and avoids any probabilistic assumption on the data, PCA has been widely used in applications [2,[5][6][7]. Our objective in this paper is to derive randomized PCA method from robust Markov chain Monte Carlo algorithm , with the aim of providing a better dimension reduction tool for real images.…”