Open Access
Properties of partial projection filter
H. Ogawa
- Vol. 71, Iss: 2, pp 527-534
About:
The article was published on 1988-01-01 and is currently open access. It has received 11 citations till now. The article focuses on the topics: Kernel adaptive filter & Filter (video).read more
Citations
More filters
Journal ArticleDOI
Consistent Sampling and Signal Recovery
Akira Hirabayashi,Michael Unser +1 more
TL;DR: This paper proposes an extension of consistent sampling that is applicable to those singular cases as well, and that yields a unique and well-defined solution that displays solutions that preserve polynomials or sinusoids, and therefore perform well in practical applications.
Journal Article
A Unified Theory of the Family of Projection Filters for Signal and Image Estimation
TL;DR: In this article, the family of projection filters, comprising the projection filter, the partial projection filter and the averaged projection filter is discussed and the unified theory of the family is discussed.
Journal Article
Mutual relations among optimum image restoration filters
TL;DR: A systematic assessment is made of the relationships between nine kinds of optimum restoration filters and it is confirmed experimentally that the parametric Wiener filter can approximate the averaged projection filter with arbitrary precision.
Proceedings ArticleDOI
A theory of over-learning in the presence of noise
K. Yamasaki,Hidemitsu Ogawa +1 more
TL;DR: Necessary and sufficient conditions for two kinds of admissibility of the rote memorization criterion by the Wiener criterion are obtained and lead to a method for choosing a training set which prevents Wiener-over-learning.
Partial Oblique Projection Learning for Optimal Generalization
Liu Benyong,Uestc Chengdu +1 more
TL;DR: This paper shows that the previously discussed S-L projection learning is inappropriate to constructing a family of projection learning, and proposes a new version called partial oblique projection (POP) learning, where a function space is decomposed into two complementary subspaces, so that functions belonging to one of the subspace can be completely estimated in noiseless case.