Dictionary-type image super-resolution system and method based on iteration projection reconstruction

A super-resolution and super-resolution reconstruction technology, which is applied in the field of dictionary image super-resolution systems, can solve the time-consuming problems of high-resolution reconstruction

Active Publication Date: 2015-11-04
PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF5 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it is very time-consuming to use the dictionary for high-resolution reconstruction block by block in this type of method

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
  • Dictionary-type image super-resolution system and method based on iteration projection reconstruction
  • Dictionary-type image super-resolution system and method based on iteration projection reconstruction
  • Dictionary-type image super-resolution system and method based on iteration projection reconstruction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0055] Such as figure 1 As shown, the present invention first provides a dictionary class image super-resolution system based on iterative projection reconstruction, which includes:

[0056] The dictionary training and projection matrix precomputing modules are used to extract high-resolution image blocks from the training image set and select corresponding low-resolution image blocks as samples from the downsampled image of the image, and then use the K-means clustering method Train the dictionary, calculate the Euclidean distance between each atom in the dictionary and all samples, select the nearest neighbor to form the K-nearest neighbor of the atom, and then use the K-nearest neig...

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 dictionary-type image super-resolution system and method based on iteration projection reconstruction. The system comprises a dictionary training and pre-calculation matrix prediction module, an iteration projection super-resolution reconstruction module based on a projection matrix and a post-processing module based on global restriction. The method comprises steps: firstly, dictionary learning and projection matrix calculation are carried out; secondly, iteration projection super-resolution reconstruction based on a projection matrix is carried out, namely, y is inputted into a low-resolution image block, an atom most similar to y is searched in a dictionary, y is subjected to super-resolution reconstruction by utilization f the projection matrix of the atom, the generated residual vector is subjected to projection reconstruction, high-resolution reconstruction of the residual vector is obtained, iteration reconstruction is carried out in the above way, finally all the reconstruction components are subjected to weighing addition and the sum is employed as the reconstruction result; thirdly, the reconstruction image is subjected to global restriction to eliminate the unnatural effect. The calculation speed of the dictionary-type method can be raised effectively and image high-frequency detail information can be recovered.

Description

technical field [0001] The invention relates to the technical field of image and video super-resolution, in particular to a dictionary-like image super-resolution system and method based on iterative projection reconstruction. technical background [0002] Super-resolution (Super-Resolution), also known as upsampling and image enlargement, refers to the recovery of high-resolution clear images from low-resolution images. Super-resolution is one of the basic problems in the field of image and video processing, and has a very broad application prospect in medical image processing, image recognition, digital photo processing, high-definition television and other fields. [0003] One of the most classic super-resolution algorithms is a kernel-based interpolation algorithm, such as bilinear interpolation, spline interpolation, and so on. However, this type of algorithm generates continuous data through known discrete data, which will bring effects such as blurring and aliasing. ...

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): G06T3/40
CPCG06T3/00G06T3/4076
Inventor 赵洋王荣刚高文王振宇王文敏
Owner PEKING UNIV SHENZHEN GRADUATE SCHOOL
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