A hyperspectral image super-resolution realization method and system

A hyperspectral image and super-resolution technology, which is applied in the field of hyperspectral image super-resolution realization method and system, can solve problems such as spectral feature loss, and achieve the effect of reducing spectral loss

Active Publication Date: 2021-05-07
SHENZHEN UNIV
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the defects in the prior art that easily cause the loss of spectral features when implementing hyperspectral image super-resolution, thereby providing a method and system for implementing hyperspectral image super-resolution

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
  • A hyperspectral image super-resolution realization method and system
  • A hyperspectral image super-resolution realization method and system
  • A hyperspectral image super-resolution realization method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] Embodiments of the present invention provide a hyperspectral image super-resolution implementation method, such as figure 1 As shown, the method includes the following steps:

[0043] Step S1: Enlarge the original hyperspectral image to a low-resolution hyperspectral image of the same size as the expected output by means of interpolation.

[0044] The embodiment of the present invention considers the low-resolution hyperspectral image of a single image, and first uses the interpolation method to enlarge it to a low-resolution hyperspectral image of the same size as the expected output. Since only spatial information is considered in the interpolation method, the interpolation process Much spatial detail information has been lost, and spectral information has also been distorted, so further processing to increase resolution is required.

[0045] Step S2: Input the low-resolution hyperspectral image into the spatial spectral feature extraction network to generate a high-...

Embodiment 2

[0066] Embodiments of the present invention provide a hyperspectral image super-resolution implementation system, such as image 3 shown, including:

[0067] The low-resolution hyperspectral image acquisition module 1 is used to amplify the original hyperspectral image to a low-resolution hyperspectral image of the same size as the expected output by means of interpolation; this module performs the step S1 in the embodiment 1. The method described will not be repeated here.

[0068] The high-resolution hyperspectral image generation module is used to input the low-resolution hyperspectral image into the spatial spectral feature extraction network to generate a high-resolution hyperspectral image. The spatial spectral feature extraction network includes: spectral mapping network, spatial The reconstruction network and the space-spectrum fusion network input the low-resolution hyperspectral images into the spectral mapping network and the spatial reconstruction network respecti...

Embodiment 3

[0071] An embodiment of the present invention provides a computer device, such as Figure 4 As shown, the device may include a processor 51 and a memory 52, wherein the processor 51 and the memory 52 may be connected via a bus or in other ways, Figure 4 Take connection via bus as an example.

[0072] As a non-transitory computer-readable storage medium, the memory 52 can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as corresponding program instructions / modules in the embodiments of the present invention. The processor 51 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory 52, that is, to realize the hyperspectral image super-resolution in the first method embodiment above. Implementation.

[0073] The memory 52 may include a program storage area and a data storage area, wherein the progra...

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 hyperspectral image super-resolution realization method and system. The method comprises the following steps: amplifying an original hyperspectral image into a low-resolution hyperspectral image with the same size as expected output in an interpolation mode; inputting the low-resolution hyperspectral image into a spatial-spectral feature extraction network to generate a high-resolution hyperspectral image, the spatial-spectral feature extraction network comprising a spectral mapping network, a spatial reconstruction network and a spatial-spectral fusion network; inputting the low-resolution hyperspectral image into the spectral mapping network and the spatial reconstruction network, performing spectral feature extraction and spatial information feature extraction respectively in the two branch networks to reconstruct spectral information and spatial information; and using a spatial-spectral fusion network to fuse the spectral information and the spatial information output by the spectral mapping network and the spatial reconstruction network to generate a high-resolution hyperspectral image. Better performance is embodied in the aspects of spatial information recovery and spectral information storage.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and system for realizing super-resolution of hyperspectral images. Background technique [0002] Hyperspectral image (HSI) has dozens or even hundreds of bands, hyperspectral image not only has rich spatial texture information, but also has rich spectral information, this feature makes it widely used in agriculture, medicine, military and remote sensing and other fields. However, due to hardware limitations of hyperspectral sensors and optical imaging systems, hyperspectral images have lower spatial resolution compared with panchromatic or multispectral images. This severely limits its further application and development. Therefore, in recent years, as one of the main technologies to improve the resolution of hyperspectral images, hyperspectral image super-resolution has received extensive attention from researchers at home and abroad. [0003] Hyperspectral i...

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): G06T3/40G06N3/04
CPCG06T3/4046G06T3/4076G06N3/04G06N3/045Y02A40/10
Inventor 李岩山陈世富周李唐浩劲谢维信
Owner SHENZHEN 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