Single image super-resolution output method based on graph model

A single-image, super-resolution technology, applied in graphics-image conversion, image data processing, instruments, etc., can solve problems such as difficulty in determining the size and type relationship of the training set, and retraining the model once

Active Publication Date: 2020-09-22
HANGZHOU DIANZI UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these traditional methods have obvious disadvantages. For example, the relationship between the size and type of the training set and the training effect is difficult to determine, and the model needs to be retrained once the sampling rate is changed.

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
  • Single image super-resolution output method based on graph model
  • Single image super-resolution output method based on graph model
  • Single image super-resolution output method based on graph model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0038] For the image block segmented in step 1, a number of blocks that are most similar to the current image block are searched through the non-local average method, and the pixel values ​​of all similar blocks in the similar block set are pulled into a vector, and then a two-dimensional image is constructed based on this vector The adjacency matrix A is used to store the connection of pixel values.

[0039] Assuming that the size of each image block is m×n, and each block finds k most similar blocks to it, then the two-dimensional adjacency matrix A is a (m×n×k)×(m×n×k)-dimensional matrix ; Carry out Laplace transform on the two-dimensional matrix A to obtain a (m×n×k)×(m×n×k)-dimensional Laplace matrix, specifically the straightened vector x and the corresponding image block The adjacency matrix A, the degree matrix D corresponding to the image block, and the Laplacian matrix L corresponding to the image block are as follows:

[0040]

[0041]

[0042] Among them, th...

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 a graph-model-based single image super-resolution output method. The graph-model-based single image super-resolution output method includes the following steps: 1) according toa preset sampling rate, performing interpolation on an input single image by means of a bicubic interpolation method, obtaining a low resolution image, segmenting the obtained low resolution image into m*n image blocks, calculating the Euclidean distance of the pixel value of each image block, and according to the Euclidean distance, determining a set of similar blocks, wherein each image block in the segmented low resolution image has one set of similar blocks; 2) constructing a graph model for the set of similar blocks; 3) performing Laplace transformation on the obtained two dimensional matrix to obtain a Laplace matrix, and then carrying out restoration of the image blocks through solution by means of an optimization formula; and 4) averaging and reconstructing the restored image block set, and finally obtaining a super-resolution image. The graph-model-based single image super-resolution output method can overcome the defect that a machine learning method needs mass image data totrain a model, thus being more suitable for image super resolution.

Description

technical field [0001] The invention relates to the field of image super-resolution, in particular to the super-resolution research of a single frame image, in particular to a single image super-resolution output method based on a graph model. Background technique [0002] In a large number of electronic imaging applications, high resolution images are often desired. High resolution means a high density of pixels in an image, providing more detail that is essential in many practical applications. For example, high-resolution medical images are very helpful for doctors to make correct diagnoses; using high-resolution satellite images it is easy to distinguish similar objects from similar ones. If high-resolution images can be provided, the performance of pattern recognition in computer vision will be greatly improved. Since the 1970s, charge-coupled devices (CCDs), CMOS image sensors have been widely used to capture digital images. Although these sensors are suitable for m...

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 Patents(China)
IPC IPC(8): G06T3/40
CPCG06T3/4007G06T3/4053
Inventor 颜成钢张腾张永兵赵崇宇李志胜
Owner HANGZHOU DIANZI 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