Journal ArticleDOI
Adaptive nonlinear prediction of TV image sequences
Reads0
Chats0
TLDR
In this paper, a scheme for interframe nonlinear adaptive prediction is proposed, where a quadratic operator, adapted pixel-by-pixel by an LMS technique, proves to be able to outperform conventional predictors when motion is present in the scene.Abstract:
A scheme for interframe nonlinear adaptive prediction is proposed. A quadratic operator, adapted pixel-by-pixel by an LMS technique, proves to be able to outperform conventional predictors when motion is present in the scene.read more
Citations
More filters
Journal ArticleDOI
Quadratic filters for signal processing
TL;DR: The principle aspects and properties of quadratic filters are derived in the framework of the discrete Volterra expansion in this article, and both fixed and adaptive filters are considered in one-dimensional and multidimensional environments.
Journal ArticleDOI
Enhancing document images with a quadratic filter
Giovanni Ramponi,Paolo Fontanot +1 more
TL;DR: Using the superposition of a linear and a quadratic operator, noise smoothing is achieved without blurring the details of the picture in gray-level images of documents.
Proceedings ArticleDOI
Adaptive predictive coding of speech by means of volterra predictors
Enzo Mumolo,D. Francescato +1 more
TL;DR: The main result is that, by using this type of predictor, lower variance error signals can be obtained, as compared to the classical, linear, case.
Journal ArticleDOI
Nonlinear System Identification: An Effective Framework Based on the Karhunen–LoÈve Transform
TL;DR: This paper proposes a general framework that is able to define a large class of nonlinear system identifiers and shows how these identifiers almost always have higher performance than that obtained by current best practices, as well as very good accuracy, optimal noise rejection, and fast algorithmic elaboration.
Journal ArticleDOI
Polynomial weighted median filtering
Kenneth E. Barner,T.C. Aysal +1 more
TL;DR: Analysis shows weighted median processing of polynomial terms is justified from a maximum likelihood perspective, and the effectiveness of PWM filters is demonstrated through simulations that include temporal, spectrum, and bispectrum analysis.