Image optimization method integrating deep learning night vision enhancement and filtering noise reduction

A technology of deep learning and optimization methods, applied in neural learning methods, image enhancement, image analysis and other directions, can solve the problems of damaged pixels, indistinct details, deterioration of night vision images, etc., to achieve good noise reduction effect and avoid over-simulation. risk of convergence and avoid the effects of over-adjustment

Pending Publication Date: 2022-02-01
FUZHOU UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the heat generated by the photosensitive sensor when it works for a long time, some pixels will be destroyed during imaging, which will make the night vision image dark overall and the details are not obvious, and will inevitably introduce a lot of noise, resulting in sharp night vision images. deterioration
[0003] Although traditional nighttime enhancement algorithms such as histogram equalization, gamma transformation, and adaptive histogram equalization can enhance nighttime images, the noise amplification effect caused by enhancement seriously affects the effect of enhanced night vision images.

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
  • Image optimization method integrating deep learning night vision enhancement and filtering noise reduction
  • Image optimization method integrating deep learning night vision enhancement and filtering noise reduction
  • Image optimization method integrating deep learning night vision enhancement and filtering noise reduction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The technical solution of the present invention will be specifically described below in conjunction with the accompanying drawings.

[0055] Such as Figure 1-6 As shown, an image optimization method that integrates deep learning night vision enhancement and filter noise reduction, uses night vision enhancement network to enhance nighttime images and uses non-local mean filter NLM to denoise the enhanced image; it is characterized in that it includes The following steps;

[0056] Step S1: Obtain images with low light and overexposure as the data set required for training;

[0057] Step S2: constructing a deep neural network for enhancing nighttime images;

[0058] Step S3: Input the original image P1 of the data set into the network constructed in step S2 to obtain the trained model M1;

[0059] Step S4: Input a low-light image P2 taken at night into the trained model to obtain an enhanced nighttime noisy image P3

[0060] Step S5: Perform non-local mean filter NLM ...

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 relates to an image optimization method integrating deep learning night vision enhancement and filtering noise reduction. The enhancement quality of the light enhancement network is evaluated through a non-reference loss function, a to-be-enhanced image is used as input, high-order curves are generated as output, then the curves are used for adjusting the input dynamic range pixel by pixel to obtain a night vision enhanced image, noise reduction processing is performed on the enhanced image by using non-local mean filtering, while the color and details of the enhanced image are kept, a large amount of noise caused by night vision enhancement is filtered out, so that the image obtains a better optimization effect. According to the invention, no obvious noise is generated under the condition that images shot at night are clear.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to an image optimization method that combines deep learning night vision enhancement and filtering noise reduction. Background technique [0002] With the increasing improvement of people's lives, the requirements for the quality of captured pictures are gradually increasing. How to better process images has become a hot research field at present, and night image enhancement technology has become the mainstream of the times. However, due to the heat generated by the photosensitive sensor when it works for a long time, it will destroy some pixels during imaging, which will make the night vision image dark overall and the details are not obvious, and will inevitably introduce a lot of noise, causing the night vision image to be sharp. deterioration. [0003] Although traditional nighttime enhancement algorithms such as histogram equalization, gamma transform, and adaptive hist...

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 Applications(China)
IPC IPC(8): G06T5/00G06T5/20G06N3/04G06N3/08
CPCG06T5/002G06T5/007G06T5/20G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/20024G06N3/045
Inventor 吴林煌王凯欣温仁芳
Owner FUZHOU UNIV
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