A gray level image noise reduction method based on hole convolution and an automatic coding and decoding neural network

A neural network, grayscale image technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve the problems of large occupation, difficult to use, large memory, etc., to achieve rapid removal, simple structure, and improved visual effects Effect

Active Publication Date: 2019-04-23
ZHEJIANG UNIV
View PDF7 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although deep neural network methods can achieve better results, such methods still need to occupy a large am

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
  • A gray level image noise reduction method based on hole convolution and an automatic coding and decoding neural network
  • A gray level image noise reduction method based on hole convolution and an automatic coding and decoding neural network
  • A gray level image noise reduction method based on hole convolution and an automatic coding and decoding neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, and do not limit the protection scope of the present invention.

[0021] In order to achieve image denoising, this embodiment provides a grayscale image denoising method based on atrous convolution and automatic codec neural network, which specifically includes building an image denoising model and using the image denoising model to denoise the noisy image two parts.

[0022] Build an image denoising model, such as Figure 4 shown, including the following process:

[0023] First prepare the training set, that is, add Gaussian noise to the clear image with a fixed noise level to obtain the noise image corresponding to the clear image,...

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 gray level image noise reduction method based on hole convolution and an automatic coding and decoding neural network, which comprises the following steps of taking a clear image and a noise image corresponding to the clear image as a training sample to construct a training set; constructing an image noise reduction model, wherein the image noise reduction model comprisesan image feature coding unit for performing feature coding on a noise image, the image feature decoding unit is used for decoding the coded feature map output by the image feature coding unit, the image feature coding unit comprises a feature extraction module and 10 coding modules, and the image decoding unit comprises 10 decoding modules and an image restoration module; utilizing the training set to train the constructed image noise reduction model to obtain a trained image noise reduction model; during application, inputting a noise image into the trained image noise reduction model, and outputting a noise reduction image through calculation. The grayscale image noise reduction method can quickly remove the noise of the image and improve the visual effect of the image.

Description

technical field [0001] The invention belongs to the field of image signal processing, and in particular relates to a grayscale image noise reduction method based on atrous convolution and automatic codec neural network. Background technique [0002] Images are an extremely important source of information for people. In today's information age, with the popularization of digital devices, digital images have become an important means for people to obtain information, penetrate into all aspects of production and life, and have achieved huge social and economic benefits. In recent years, the combination of image processing technology and research fields such as machine learning and machine vision has produced unprecedented new developments and breakthroughs. However, in the process of image acquisition, processing, compression, transmission, storage and reproduction, noise will inevitably be introduced, thereby reducing image quality. The main goal of image denoising is to fil...

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): G06T5/00G06N3/04G06N3/08
CPCG06N3/084G06T5/002G06T2207/20081G06T2207/20084G06N3/045
Inventor 陈耀武李圣昱周凡
Owner ZHEJIANG 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