Home // CONTENT 2019, The Eleventh International Conference on Creative Content Technologies // View article

Real-Time Noise Level Detection for General Video

Authors:
Chinatsu Mori
Seiichi Gohshi

Keywords: Video noise reducer; 4KTV; 8KTV; Real-time; Nonlinear signal processing; Image quality.

Abstract:
Currently, 4K TV is a standard TV and 8K broadcasting has began in December 2018. High-resolution in conjunction with low noise is an essential figure of merit in video systems. Unfortunately, any increase in resolution unavoidably increases noise levels. A signal processing method called noise reducer (NR) is often used to reduce noise. However, accurate noise level is needed when NR is employed. Noise level depends mostly on lighting conditions and is estimated by comparing adjacent frame difference. However, the frame difference is generated by moving objects as well as noise. Therefore, it is essential to determine whether the frame difference is caused by the moving objects or by the noise, which is a difficult task. Another difficulty arises from the fact that noise level detection must be achievable in realtime conditions since all video systems are required to work in real-time. This means that a complex method could not be used for noise level detection. In this paper, two noise level detection algorithms are presented. The combination of two of them is a concise algorithm able to accurately detect the noise level and work in real-time conditions.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2019

Publication date: May 5, 2019

Published in: conference

ISSN: 2308-4162

ISBN: 978-1-61208-707-8

Location: Venice, Italy

Dates: from May 5, 2019 to May 9, 2019