Video denoising processing method and device based on fixed scene

A technology for video noise reduction and noise reduction processing, which is applied to TVs, color TVs, components of color TVs, etc., and can solve problems such as noise reduction and misjudgment.

Active Publication Date: 2013-12-04
SHENZHEN ZTE NETVIEW TECH
View PDF8 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] In order to overcome the defect of noise reduction misjudgment existing in existing noise reduction methods, the p

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video denoising processing method and device based on fixed scene
  • Video denoising processing method and device based on fixed scene
  • Video denoising processing method and device based on fixed scene

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0096] In a preferred embodiment, the processing module 300 performs background modeling on the down-sampled key component video data to extract the foreground area and the background area. The method includes:

[0097] The state of each pixel point in the nth frame uses a single Gaussian distribution model η(X n ,μ n ,∑ n ) to represent, where μ n is the mean value of the pixel at the nth frame, ∑ n is the covariance of the pixel at the nth frame, if the feature factor greater than the preset judgment threshold T, it is judged that the pixel is a foreground pixel; otherwise, the feature value of the pixel is considered to match the Gaussian distribution, and the pixel is judged to be a background pixel, wherein the d n =I n -μ n , I n is the gray value of the current pixel.

[0098] Under a more preferred implementation manner, the processing module 300 refers to the single Gaussian distribution model η (X n ,μ n ,∑ n ) to update:

[0099] mu n+1 =(1-α)μ n +αI ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a video denoising processing method and device based on a fixed scene. The method includes the steps of obtaining a frame of video data, extracting key component video data in the frame of the video data, performing downsampling processing on the key component video data, performing background modeling on the downsampled key component video data so as to extract a foreground region and a background region, estimating a current noise power value of the background region according to the difference value of the current pixel value of the background region and the pixel value of a prior background region, performing no denoising processing on the key component video data if the noise power value is less than a preset denoising processing judgment threshold, and performing denoisng processing on the foreground region and the background region of the key component video data if the noise power value is larger than the preset denoising processing judgment threshold. According to the video denoising processing method and device, the subjective quality of videos can be improved, calculation amount and storage resources are low, and hardware implementation is easy.

Description

technical field [0001] The present invention relates to the field of video monitoring, in particular to a video noise reduction processing method and device based on a fixed scene. Background technique [0002] At present, in the video surveillance system, it is generally required to monitor 24 hours a day, but in practical applications, the video collected by the sensor inevitably has noise, especially in the low-light environment at night or in the evening, there will be noise in the video. Lots of noise. There are many sources of these noises, which can be summarized as follows: [0003] 1. Noise introduced by the front-end sensor used to collect video; [0004] 2. Noise introduced due to electromagnetic interference during video transmission; [0005] 3. During video compression, the noise introduced by the quantization process, etc. [0006] The introduction of these noises not only seriously reduces the subjective quality of video, but also reduces the accuracy of ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): H04N5/21
Inventor 张雪林邓海波吴贻刚朱豪
Owner SHENZHEN ZTE NETVIEW TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap