Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image high and low frequency decomposition noise removal method based on residual dense network

A dense network and noise decomposition technology, applied in the field of image processing, can solve the problems of blurred edge information, easy to generate artifacts, etc., and achieve the effect of thorough removal and ideal results

Pending Publication Date: 2021-06-18
XIAN UNIV OF TECH
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a high and low frequency decomposition noise removal method based on the residual dense network, which solves the problems of blurred edge information and easy generation of artifacts in the prior art after image denoising

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 high and low frequency decomposition noise removal method based on residual dense network
  • Image high and low frequency decomposition noise removal method based on residual dense network
  • Image high and low frequency decomposition noise removal method based on residual dense network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0050] The present invention is a high and low frequency decomposition noise removal method based on the residual dense network, such as figure 1 As shown, it specifically includes the following steps:

[0051] Step 1, obtain the image to be trained;

[0052] The specific process of step 1 is:

[0053] Select A training sample images to form the training sample image set, randomly select m images from the training sample image set, randomly cut out i pictures with size n×n from each picture, and obtain m×i pictures with size n× n pictures.

[0054]Step 2, add Gaussian noise to the image to be trained in step 1, and construct an image pair X;

[0055] The specific process of step 2 is:

[0056] Step 2.1, copy the m×i pictures processed in step 1, and add the same Gaussian noise to each copied image to obtain an artificial noise picture; ...

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 an image high and low frequency decomposition noise removal method based on a residual dense network. The method specifically comprises the following steps: step 1, obtaining a to-be-trained image; 2, adding Gaussian noise to the to-be-trained image in the step 1, and constructing an image pair X; 3, obtaining a corresponding high-frequency image training sample and a corresponding low-frequency image training sample from the to-be-trained image in the step 1 and the noise data set obtained in the step 2 by using a high-pass filter respectively; 4, respectively inputting the high-frequency image training sample and the low-frequency image training sample obtained in the step 3 into a residual dense network for training, and respectively obtaining a high-frequency image subjected to noise removal and a low-frequency image subjected to noise removal; 5, adding the denoised high-frequency image and the denoised low-frequency image obtained in the step 4 in a one-to-one correspondence manner to obtain a denoised overall image. The problems that in the prior art, after image denoising, edge information is fuzzy, and artifacts are likely to be generated are solved.

Description

technical field [0001] The invention belongs to the technical field of image processing methods, and relates to an image high and low frequency decomposition noise removal method based on residual dense network. Background technique [0002] Due to the influence of various factors such as imaging equipment, the image will be disturbed by noise during the imaging or sensing process, which will affect subsequent tasks such as image segmentation and target recognition and cannot be carried out smoothly. For example, when public monitoring equipment is used to identify criminal suspects, the noise of the image makes it very difficult to identify the facial features of the criminal suspect; in remote sensing images, small target objects have fewer imaging pixels, and the presence of noise makes it difficult to identify small target objects in the image. In view of the above situations, how to accurately remove image noise and at the same time protect the original image details fr...

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): G06T5/00G06N3/08
CPCG06N3/08G06T2207/20081G06T2207/20084G06T5/70
Inventor 刘晶向朋霞何帅
Owner XIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com