Hyperspectral image reconstruction method adopting deep learning

A hyperspectral image and deep learning technology, applied in the field of reconstructing hyperspectral images using deep learning, can solve the problem of ignoring the spectral details of the scene

Inactive Publication Date: 2017-08-01
HANGZHOU DIANZI UNIV
View PDF3 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, doctors often need to obtain image information of a certain part of the patient's body to conduct a more in-depth pathological analysis. At this time, ordinary cameras cannot meet this requir

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
  • Hyperspectral image reconstruction method adopting deep learning
  • Hyperspectral image reconstruction method adopting deep learning
  • Hyperspectral image reconstruction method adopting deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0033] The present invention proposes to use the existing hyperspectral image to establish a deep neural network, and apply it to the implementation process of restoring the existing two-dimensional sensor image into a hyperspectral image. figure 1 shown. The method of the present invention comprises the following steps:

[0034] Step (1) Obtain a hyperspectral dataset.

[0035] First put the sample hyperspectral images that need to be trained in the specified folder, then convert each sample hyperspectral image into a single-channel image, and then cut the single-channel image into small images of 20 pixels*20 pixels, so that A hyperspectral dataset is obtained; it is convenient for the next steps to operate on it.

[0036] Step (2) train and obtain the neural network by means of sparse coding.

[0037] The images in the hyperspectral dat...

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 reconstruction method adopting deep learning. The hyperspectral image reconstruction method comprises steps that 1) a hyperspectral training data set is trained by adopting a sparse coding way to acquire a neural network; 2) space spectrum modulation of an original hyperspectral image of a target scene is carried out to acquire a two-dimensional sensor image; 3) the two-dimension sensor image acquired in the step 2) is reconstructed by using the neural network, and the reconstructed hyperspectral image of the target scene is acquired. The hyperspectral image reconstruction method is advantageous in that a problem of low hyperspectral imaging speed is solved, and the hyperspectral image having higher resolution is acquired.

Description

technical field [0001] The invention relates to the field of hyperspectral image reconstruction, in particular to a method for reconstructing hyperspectral images using deep learning. Background technique [0002] In today's era, ordinary cameras can no longer meet people's requirements in some fields. In actual work and production, it is often necessary to obtain actual information through higher-resolution images. For example, doctors often need to obtain the image information of a certain part of the patient's body to conduct a more in-depth pathological analysis. At this time, ordinary cameras cannot meet this requirement. However, in related technologies, a color camera obtains color records of a scene by adding red, green, and blue filters to the sensor, which often ignores the spectral details of the scene. Therefore, in order to collect three-dimensional hyperspectral images, time-sharing scanning is often adopted (that is, by sacrificing time resolution or spatial...

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): G06T3/40G06T7/90G06N3/04
Inventor 姜鑫颜成钢吴嘉敏吴桐崔恩楠彭冬亮薛安克
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