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

Rain removal method based on feature supervision generative adversarial network

A network and generator technology, applied in the field of image processing, can solve problems such as blurring and covering, achieve the effect of strengthening feature propagation, alleviating the problem of vanishing gradients, and improving the flow of gradients

Pending Publication Date: 2021-03-16
NANCHANG HANGKONG UNIVERSITY
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Under rainy conditions, the obtained image or video will usually have some bad effects such as blurring and covering due to the interference of rain

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
  • Rain removal method based on feature supervision generative adversarial network
  • Rain removal method based on feature supervision generative adversarial network
  • Rain removal method based on feature supervision generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] In the present invention, one or more of the following terms are used.

[0026] Convolution (Convolution, Conv): When the convolution kernel is image processing, the weighted average of pixels in a small area of ​​​​the input image becomes each corresponding pixel in the output image, where the weight is defined by a function, which is called volume Accumulation.

[0027] Batch Normalization (Batch Normalization, BN): BN performs a similar standardization operation on the input value or the tensor of the convolutional network, and scales it to an appropriate range, thus speeding up the training; on the other hand, each layer can be Try to face the input values ​​of the same characteristic distribution, reducing the uncertainty caused by the change.

[0028] Activation function: The activation function is very important for the artificial neural network model to learn and understand very complex and nonlinear functions. It refers to how to retain and map the "characteri...

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 the technical field of image processing, and provides a rain removal method based on a feature supervision generative adversarial network. In order to improve the robustness and the parameter utilization rate of the network, a Dense Block module is used as a constituent part of a U-Net network structure in a generator. Each convolution layer of the DenseNet network structure can be connected with other convolution layers, so that feature propagation is enhanced, and the utilization rate of parameters is increased. Besides, the activation function of the hidden layer ofthe network adopts Leaky ReLu to replace ReLU so as to solve the problem that when the input value of the ReLU is negative, the output is always 0, neurons cannot update parameters, namely the neurons do not learn.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for removing rain from an image. Background technique [0002] Under rainy conditions, the obtained image or video usually has some bad effects such as blurring and covering due to the interference of rain. Therefore, it is a very meaningful subject to study image deraining technology. [0003] Existing deraining methods are mainly divided into two categories: single image-based and video-based deraining methods. In video-based deraining algorithms, the usual method is to use additional time information. In the method of removing rain based on a single image, the method based on the prior usually leads to excessive smoothing of the background. In recent years, in addition to the method based on the prior, people usually use the convolutional neural network method. Experiments have proved that the convolutional neural network The method of the network performs...

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): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/00G06V10/40G06N3/045G06F18/214
Inventor 盖杉卢贝
Owner NANCHANG HANGKONG UNIVERSITY
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