Unlock instant, AI-driven research and patent intelligence for your innovation.

MR image super-resolution method

A super-resolution and image technology, applied in the field of computer vision and image processing, can solve the problem of low quality of super-resolution results and achieve excellent performance

Active Publication Date: 2021-06-11
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The main purpose of the present invention is to propose a method for super-resolution of MR images based on a pyramidal orthogonal attention network based on double self-similarity, so as to solve the technical problem that the quality of super-resolution results in the above-mentioned prior art is low

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
  • MR image super-resolution method
  • MR image super-resolution method
  • MR image super-resolution method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] The present patent application will be described in further detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be emphasized that the following descriptions are only exemplary and not intended to limit the scope and application of this patent application.

[0028] Non-limiting and non-exclusive embodiments will be described with reference to the following drawings, wherein like reference numerals refer to like parts unless specifically stated otherwise.

[0029] Those skilled in the art will recognize that many variations on the above description are possible, so the examples are merely intended to describe one or more particular implementations.

[0030] Embodiments of the present invention provide a method for super-resolution of MR images based on a dual self-similarity pyramidal orthogonal attention network, comprising the following steps:

[0031] The first step: establish a pyramid dual self-similarity...

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 an MR image super-resolution method, which comprises the following steps of: 1, establishing a pyramid double-self-similarity module (PDSB), simultaneously fusing global point correlation and pyramid correlation containing multi-scale block correlation by the module, and fully extracting self-similarity prior characteristics of an MR image in a spatial domain by using the PDSB; 2, establishing a non-dimensionality reduction channel attention module NCAB in a channel domain; 3, combining the PDSB of the spatial domain and the NCAB of the channel domain into a pyramid orthogonal attention module POAB, and combining the pyramid orthogonal attention module POAB with a plurality of residual modules RB and a pixel up-sampling layer to form a pyramid orthogonal attention network POAN for MR image super-resolution; and 4, performing super-resolution on the input MR image by using the trained network suitable for different super-resolution scales to obtain an output high-definition image. By means of the MR image super-resolution method, the high-quality MR image super-resolution effect can be achieved.

Description

technical field [0001] The invention relates to the fields of computer vision and image processing, in particular to an MR image super-resolution method. Background technique [0002] The use of medical imaging to assist diagnosis and treatment has become an indispensable detection item in current medical detection. Magnetic resonance images (MR) can reflect abnormalities and early lesions of human organs from inside the molecules of the human body, and have good imaging and diagnostic effects in the brain, heart and other parts. Compared with other types of medical images, high-quality MR images have higher resolution and can contain detailed structural and tissue information, so they are more suitable for clinical diagnosis, decision-making, and accurate quantitative image analysis. However, due to hardware and physical equipment limitations, high-resolution imaging comes at the cost of long scan times, small spatial coverage, and slow spatial convergence. In addition, t...

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
IPC IPC(8): G06T3/40G06N3/04G06N3/08G06K9/46G06K9/62
CPCG06T3/4053G06T3/4046G06N3/08G06V10/44G06N3/047G06N3/045G06F18/22
Inventor 王好谦胡小婉
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV