Picture quality enhancement system and method based on meta learning, and storage medium

An enhanced system and meta-learning technology, applied in the field of image processing, can solve problems such as unsatisfactory results and lack of excellent algorithms, and achieve the effects of good universality, reduced training complexity, and improved efficiency.

Inactive Publication Date: 2021-09-28
上海人工智能研究院有限公司
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is no excellent algorithm for the fusion of the denoising task and other tasks, and the results of some algorithms for the fusion of the three tasks are not satisfactory.

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
  • Picture quality enhancement system and method based on meta learning, and storage medium
  • Picture quality enhancement system and method based on meta learning, and storage medium
  • Picture quality enhancement system and method based on meta learning, and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] The basic principles of metamorphisms involved in this embodiment are as follows: Yuan learning or learning learning (Learn Tolearn) has become another important research branch after reinforcement learning. A good machine learning model typically requires training using a large number of samples. In contrast, human beings can learn new concepts and skills faster and more effectively. Through some training a small number of samples, a machine learning model with similar properties is quickly designed, which is a problem that is aimed at solving. The method of meta-learning is diverse, there is a method based on gradient prediction, and there is also a memory-based approach, and there is a method based on attention mechanism.

[0044] like figure 1 Distance figure 1 For the overall frame structure of the system, the system provided by this embodiment includes a denoising processing network sequentially connected, a deprived processing network, a super-processing network;

[...

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 quality enhancement system and method based on meta learning, and a storage medium. The system comprises a denoising processing network, a deblurring processing network and a super-resolution processing network which are connected in sequence. The denoising processing network is used for de-noising an input image; the deblurring processing network is used for deblurring the input image; and the super-resolution processing network is used for performing super-resolution processing on the input image. According to the system, the image enhancement technology is integrated, the joint task of denoising, deblurring and super-resolution is realized, and through a meta transfer learning deblurring algorithm, the training speed of the deblurring network is increased, the training complexity is reduced, and the image enhancement network has better universality. According to the method, important information in the image is highlighted according to specific needs, unnecessary information is weakened or removed, the efficiency of network training is improved through meta learning, and the effect of the network can be improved.

Description

Technical field [0001] The present invention relates to the field of image processing, particularly a metallic enhancement method and system based on meta-learning, storage medium. Background technique [0002] Because of the problem of hand shaking or focus choice, there is often a blurred condition in the image captured by the camera. At the same time, low resolution is a problem that security video and old video is easy to discover. The picture noise is a problem that the image quality decline during image transportation. These are the basic issues of computer vision and image processing. These issues affect the industry of high quality needs, for example, clear medical images is very helpful for doctors; use high resolution The noise satellite image is easy to distinguish from similar objects from the similar object; if it is possible to provide a clearer image, the information contained in the image is more, and the performance of the pattern identified in the computer visio...

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): G06T5/00G06T3/40G06N3/04G06N3/08
CPCG06T5/002G06T5/003G06T3/4007G06T3/4053G06T3/4046G06N3/08G06T2207/10016G06T2207/20081G06T2207/20084G06N3/048G06N3/045
Inventor 宋海涛盛斌王资凯沈灏王天逸石嵘昱李佳佳章笑晨
Owner 上海人工智能研究院有限公司
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