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

Multi-resolution feature cross fusion image compressed sensing reconstruction system and method thereof

A high-resolution image and multi-resolution technology, applied in the field of image imaging and reconstruction, can solve the problems of limiting the feature expression ability of deep network and system reconstruction performance

Active Publication Date: 2021-07-30
SOUTH CENTRAL UNIVERSITY FOR NATIONALITIES
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, existing schemes are basically based on multi-scale feature enhancement within a single resolution image, thus limiting the feature expression ability and system reconstruction performance of deep networks

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
  • Multi-resolution feature cross fusion image compressed sensing reconstruction system and method thereof
  • Multi-resolution feature cross fusion image compressed sensing reconstruction system and method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] Below in conjunction with accompanying drawing and embodiment describe in detail:

[0071] 1. System

[0072] 1. Overall

[0073] Such as figure 1 , the system includes the input image measurement value Y, and is provided with a multi-resolution initial reconstruction module 10 and a multi-resolution enhanced reconstruction module 20, and the system output is a reconstructed high-resolution image X H and the reconstructed low-resolution image X L ;

[0074] The multi-resolution initial reconstruction module 10 is provided with a parallel high-resolution initial reconstruction unit 11 and a low-resolution initial reconstruction unit 12, and the multi-resolution enhanced reconstruction module 20 is provided with a depth feature extraction unit 21, a depth feature extraction 2 unit 22, channel attention unit 23, feature discrimination enhancement 1 unit 24, feature discrimination enhancement 2 unit 25, feature cross fusion unit 26, high resolution residual reconstructi...

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 multi-resolution feature cross fusion image compressed sensing reconstruction system and a method thereof, and relates to the technical field of image imaging and reconstruction. The system comprises an input image measurement value Y, is provided with a multi-resolution initial reconstruction module (10) and a multi-resolution enhanced reconstruction module (20), and outputs a reconstructed high-resolution image XH and a reconstructed low-resolution image XL. The method comprises the following steps: (1) performing multi-resolution initial reconstruction; (2) performing multi-resolution enhancement reconstruction; and (3) training the reconstruction system. Compared with the prior art, the method has the advantages that the reconstruction performance of the system can be further improved, and meanwhile, high-quality reconstruction of images with multiple resolutions is obtained; so that the method is suitable for compression imaging.

Description

technical field [0001] The present invention relates to the technical field of image imaging and reconstruction, in particular to an image compression sensing reconstruction system and its method based on cross-fusion of multi-resolution features; in detail, the present invention is a joint optimization framework based on measurement matrix and depth reconstruction , a compressed sensing reconstruction system and method for improving image restoration quality using multi-resolution feature cross-fusion technology. Background technique [0002] Compressed sensing is a new theory of signal sampling. Compared with traditional Shannon sampling, compressed sensing proves that accurate reconstruction of sparse signal dimensionality reduction sampling can be achieved, so the amount of sampled data can be greatly reduced, thereby effectively reducing data storage space and data transmission bandwidth. Compressed sensing has been widely used in image imaging. Image compression sens...

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): G06T3/40G06T5/50G06N3/04G06N3/08
CPCG06T3/4053G06T5/50G06N3/08G06T2207/20081G06T2207/20084G06T2207/20221G06N3/045
Inventor 熊承义秦鹏飞高志荣刘川鄂
Owner SOUTH CENTRAL UNIVERSITY FOR NATIONALITIES