An Adaptive Noise Reduction Method for Low-light Video Images Based on Gradient-Guided Filtering

A video image and guided filtering technology, which is applied in the field of image processing, can solve problems such as large amount of computation, lower image quality, and complex algorithms, and achieve the effects of improving the effect of filtering and noise reduction, improving image quality, and improving signal-to-noise ratio

Active Publication Date: 2020-08-11
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The human eye is very sensitive to noise, and a large amount of noise will seriously reduce the imaging quality and affect the normal observation of people. Therefore, how to effectively reduce the noise in night vision low-light imaging has always been a key technology in the research of night vision imaging technology.
[0004] Typical simple methods of image noise reduction mainly include: mean filter, median filter, bilateral filter, etc. Although the method is simple and easy to implement, the effect is often poor; typical complex methods mainly include: non-local mean (NLMS) method, Bay Yaesian least squares-Gaussian scale mixture method (BLS-GSM), three-dimensional block matching (BM3D) method, etc., although the effect is good, but the algorithm is complex, the amount of calculation is large, and it is difficult to apply in real time
[0005] In summary, how to efficiently reduce noise in low-light video images, and that can run fast and in real-time, remains a challenging problem

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
  • An Adaptive Noise Reduction Method for Low-light Video Images Based on Gradient-Guided Filtering
  • An Adaptive Noise Reduction Method for Low-light Video Images Based on Gradient-Guided Filtering
  • An Adaptive Noise Reduction Method for Low-light Video Images Based on Gradient-Guided Filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] Such as figure 1 As shown, a gradient-guided filter-based adaptive noise reduction method for low-light video images disclosed in this embodiment includes the following steps:

[0065] Step 1, such as figure 2 As shown, the motion region in the low-light video image is detected.

[0066] Step 1.1: Perform guided filtering on the current low-light video image g(x) of the xth frame and the previous low-light video image g(x-1) respectively, and use the input image to be filtered as the guiding image, and obtain the current The low-frequency information g of the frame and the previous frame of the video image L (x) and g L (x-1). Wherein, the filtering radius r=6, and the filtering smoothing coefficient λ=0.4.

[0067] Step 1.2, obtain the low-frequency image information g of the current frame L (x) and the previous frame of low-frequency image information g L The difference between (x-1), and take the absolute value, according to the formula (1) to get the low-fre...

Embodiment 2

[0085] Such as figure 1 As shown, a method for adaptive noise reduction of low-light video images based on gradient-guided filtering disclosed in this embodiment includes detecting motion areas of low-light video images, estimating the noise of low-light video images, and then performing adaptive gradient Guided filtering and iterative guided filtering processing to obtain the final filtered output image. The signal-to-noise ratio of the low-light video image after the noise reduction processing in this embodiment is significantly improved, and the image quality is significantly improved. At the same time, almost no human intervention is required, and adaptive noise reduction can be quickly realized, which has a good practical application prospect.

[0086] figure 2 A flow chart of a method for detecting a motion region of a low-light video image according to this embodiment is shown. First, input the current xth frame and the previous low-light video image g(x) and g(x-1) ...

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 low-light video image self-adaptive noise reduction method based on gradient guide filtering, and belongs to the technical field of image processing. The method comprises thefollowing steps: step 1, detecting a motion region in a low-light video image; step 2, estimating the noise intensity of the low-light video image; step 3, carrying out self-adaptive gradient guide filtering and iterative guide filtering processing on the low-light video image, so that a final filtered output image is obtained. The invention aims to provide a low-light video image self-adaptive noise reduction method based on gradient guide filtering, on the premise that human intervention is reduced as much as possible, the image noise is reduced, the image signal-to-noise ratio is improved,and the image quality is improved. The method has a good practical application prospect in the field of low-light night vision imaging.

Description

technical field [0001] The invention relates to a gradient-guided filter-based self-adaptive noise reduction method for low-light video images, which belongs to the technical field of image processing. Background technique [0002] Low-light night vision imaging is one of the important methods commonly used in the field of night vision. The inherent random noise of low-light night vision images is a key factor affecting the imaging quality. Low-light night vision image filtering has also become a long-term research topic at home and abroad. [0003] At present, the main low-light imaging devices for night vision include: image tube coupled CCD (ICCD), electron multiplier CCD (EMCCD), electron bombardment CCD (EBCCD), high-sensitivity CMOS, etc. Since the intensity of visible light at night is relatively weak, not only high-sensitivity photosensitive devices are required, but also in order to obtain as much scene detail information as possible, the photoelectric conversion si...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/00G06T5/10G06T7/254G06T7/269
Inventor 李力朱进金伟其韩正昊
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
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