EBU Logo EBU Logo
0047B - Video Noise
published
0047B v2.0
Item image
VIDEO NOISE
Visible Video Noise
EBU Tech 3335, ...
Noise below threshold
Noisy Segments, Statistics
https://qc.ebu.io/items/0047B/versions/2-0-0/
Check
Report
Summary
EBU QC Id
0047B
Name
Video Noise
Status
published
Category
subjective
Lead editor
Peter Schallauer
Aliases
image noise noise
Tags
video
Versions
Version Status Edition date Saved by Note
Details
Description
Noise constitutes an unwanted signal that inevitably adds to the useful part of the signal. Noise may originate from different sources, e.g. electronic sensor noise, quantisation noise, film grain noise etc.
Definition
Used in Checks, the system shall detect video segments showing a noise level that is above a user-defined threshold. Optionally the system may report a video noise level measurement over time.

Used in Reports, the system shall provide a video noise level measurement over time.
References

  • [https://tech.ebu.ch/docs/tech/tech3335.pdf EBU Tech 3335]

    Remark: This document gives guidance on noise levels in cameras (provided as guidance).

Examples

  • Input: "Video Noise Threshold" = 30 dB
    Input: "Video Noise Minimum Segment Duration" = 1 second
    Output: "Check Result" = false
    Output: "Video Noise Segment" from 00:00:01:00 to 00:00:02:21
    --> "Noise Average" = 32.1 dB
    --> "Noise Maximum" = 35 .2 dB

    Remark: Used as a Check. The reported Noise Average of 32.1 dB is beyond the specified threshold.

  • Input: "Video Noise Measure Interval" = 5.0 seconds
    Output: "Video Noise Level Measured Over Time" = 29.1 dB 29.2 dB 28.9 dB 28.4 dB 28.6 dB ...

    Remark: Used as a Report. The output is formed by the measured values separated by spaces.

  • uploads/items/example_images/0047Bex001.jpg

Input parameters
Name Default value Description Data model name Unit Locator
Output parameters
Name Alias Description Data model name Unit Locator
Test material
Title Version Download
Remarks
Remarks
The implementation of video noise detection is vendor specific. The noise level value range and unit is implementation dependent. This applies to the detection algorithms deployed and the thresholds to trigger reporting.

Although it is possible to estimate the visual noise level by a signal to noise ratio (SNR) measure, in a finished programme the results often need interpretation by an Operator (Golden Eyes).
User discussions
Title Author Last update