Image color cast removal method of generative adversarial network

A generative, image technology, applied in biological neural network models, image enhancement, image analysis, etc., to achieve excellent processing performance, improved color cast removal ability, and good stability.

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

AI Technical Summary

Problems solved by technology

[0008] At the problem of existing research method, the present invention research content comprises the following several parts:

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 color cast removal method of generative adversarial network
  • Image color cast removal method of generative adversarial network
  • Image color cast removal method of generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] The technical solution of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0052] The present invention can use a computer to train and infer the network, and realize it using the Tensorflow deep learning framework under the windows operating system. The specific experimental environment configuration is as follows:

[0053] GPU NVIDIA GeForce RTX 2060SUPER CPU AMD Ryzen 5 3600 6-Core Processor Programming language Python 3.7 Deep Learning Framework TensorFlow-GPU 2.1.0

[0054] The present invention mainly includes two parts: (1) using the SFU Grayball data set used for color constant calculation as a training set, and adopting the method of artificially adding simulated complex light sources, expanding this data set to be suitable for training uniform light sources, The data set required for the experiments of the color constancy model of non-uniform light sources.

[0055] ...

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 a color cast removal model based on a generative adversarial network, which is used for solving the problem that a traditional color constancy algorithm cannot process the image color cast under a complex light source. The model is expanded on the basis of a core viewpoint of an end-to-end model of a conditional generative adversarial nets (CGAN), the U-Net connection is adopted as a generator network structure, a Markov discriminator serves as a discriminator network structure, a loss function capable of better reflecting the tone information is introduced to optimize a training model, and finally an image with color cast removed is obtained. According to the method, images with inconsistent input color projection can be received, compared with the existing method, two tasks of color cast removal and image resolution improvement under any light source can be completed at the same time, and a better enhancement effect can be achieved.

Description

technical field [0001] The invention belongs to computer vision enhancement technology, in particular to an image enhancement method based on a generative confrontation network for color shift removal. Background technique [0002] In the field of computer vision research, image understanding is one of its important research directions, and the basis of image understanding is image features, in which color is one of the most important underlying features of images, and has great potential in image feature extraction and feature learning. strong reference. However, the color information of objects in image scenes is unstable, and the color of objects is easily affected by changes in light sources. The purpose of removing the color cast of the image is to perceive the color correctly and obtain the real color information of the object that has nothing to do with the change of the light source. [0003] Since the color constancy of the color image itself is limited and compli...

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): G06T5/00G06K9/62G06N3/04G06N3/08G06V10/774G06V10/82
CPCG06T5/009G06T5/003G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06N3/045G06F18/214
Inventor 曹丽琴李治江宋争光杨鹏金佳惠
Owner WUHAN 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