ISBN 978-3-8439-2010-0

978-3-8439-2010-0, Reihe Ingenieurwissenschaften

Eugen Wige
Methods for Improved Video Compression Using Image Noise Modeling

277 Seiten, Dissertation Universität Erlangen-Nürnberg (2014), Softcover, A5

Zusammenfassung / Abstract

The aim of this work is to find new approaches to improve the efficiency of video signal compression at high quality and even for lossless compression. Especially at these quality levels the inherent noise of the video signal dominates the compression performance. To improve the capabilities of conventional video codecs for such scenarios, a signal model is established where the noise is considered as an important part of the signal and has to be preserved.

Based on these assumptions, on the one hand, the state-of-the-art coding techniques are analyzed in terms of their response to noisy input signals. On the other hand, prediction techniques are developed which are less sensitive to the noise of the input video signals, but still exploit their inherent redundancy.