Supercharge Your Innovation With Domain-Expert AI Agents!

Image motion blur blind removal method based on generative adversarial network

A technology for motion blur and image generation, applied in biological neural network models, image enhancement, image analysis, etc., can solve problems such as ignoring pixel differences and correlations, limiting network feature expression and learning capabilities, and achieving a large receptive field Effect

Pending Publication Date: 2022-08-02
YANSHAN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The information of different spatial locations in the feature map is treated indiscriminately, ignoring the spatial pixel differences and correlations in a larger range, which also limits the feature expression and learning capabilities of the network to a certain extent.

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 motion blur blind removal method based on generative adversarial network
  • Image motion blur blind removal method based on generative adversarial network
  • Image motion blur blind removal method based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention will be described in detail below with reference to the accompanying drawings.

[0031] A method for blindly removing image motion blur based on generative adversarial network proposed by the present invention is implemented according to the following steps:

[0032] Step 1. Build the structure of the generative adversarial network, as follows:

[0033] The generative adversarial network includes a generator and a discriminator, such as figure 1 As shown, in the generator, 7×7 convolution and two 3×3 convolutional layers with stride 2 are used to downsample the image, and then 9 residual modules are used to extract features, each residual module It consists of a 3×3 convolutional layer, an instance normalization layer, a ReLU activation layer, a Dropout layer with a probability of 0.5, a 3×3 convolutional layer, and an instance normalization layer in turn. The residual module cascades two cross-attention modules to adaptively integrate local feat...

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 provides an image motion blur blind removal method based on a generative adversarial network. According to the method, firstly, a space attention module is added into a generator, rich context information among all pixels is utilized to implement geometric constraint on a target structure in a restored image, and global information is introduced to reduce information loss; secondly, a local-global dual-scale discriminator is adopted to consider local texture and global structure information of the image; and finally, designing a multi-component loss function, and carrying out image high-frequency edge and detail reconstruction in combination with the constraint model. According to the method, a single-image blind motion blur removing method is improved, and an image with a more obvious structure and richer details can be effectively restored from a non-uniform motion blur image. The method has wide application prospects in the fields of target tracking, traffic detection, medical imaging, military reconnaissance and the like.

Description

technical field [0001] The invention relates to the technical field of digital image processing, in particular to a method for blindly removing image motion blur based on a generative confrontation network. Background technique [0002] Motion blur is the most common type of image blur, and the removal of image motion blur has always been a popular research direction in the field of image deblurring. The main cause of motion blur is rapid relative motion between the device and the subject during the exposure time. Blurred images not only degrade human perceptual quality, but also negatively impact the accuracy of algorithms for advanced vision tasks such as object detection and semantic understanding. Therefore, image deblurring has always been a basic but very important problem in the field of image processing, and has important academic and application values. [0003] Image deblurring algorithms can be divided into two categories: non-blind deblurring and blind deblurri...

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): G06T5/00G06T7/13G06N3/08G06N3/04
CPCG06T7/13G06N3/084G06T2207/20081G06N3/045G06T5/73
Inventor 张玉存李涛米松涛
Owner YANSHAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com