Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Hyperspectral remote sensing image classification method and device

A technology for hyperspectral remote sensing and image classification, which is applied in the field of hyperspectral remote sensing image classification methods and devices, can solve problems such as decreasing and deepening accuracy, and achieves the effects of large data discrimination, good accuracy, and suppression of useless information

Pending Publication Date: 2021-03-30
HOHAI UNIV
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a hyperspectral remote sensing image classification method and device, which can solve the technical problem of "decreasing precision" caused by deepening the network depth in the prior art

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 remote sensing image classification method and device
  • Hyperspectral remote sensing image classification method and device
  • Hyperspectral remote sensing image classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] This embodiment discloses a hyperspectral remote sensing image classification method, which is a hyperspectral remote sensing image classification method based on attention and dense residual network (ADRN), such as figure 1 shown, including the following steps:

[0051] Step S1: Input the original hyperspectral remote sensing image A∈R H×W×B , where H represents the height of the hyperspectral remote sensing image, W represents the width of the hyperspectral remote sensing image, B represents the number of original bands of the hyperspectral remote sensing image, R H×W×B Representing a real number set whose size is H×W×B, and performing normalization processing on the original hyperspectral remote sensing image to obtain a normalized hyperspectral remote sensing image;

[0052] Step S2: Perform dimensionality reduction processing on the normalized hyperspectral remote sensing image by principal component analysis (PCA) to obtain hyperspectral remote sensing image data...

Embodiment 2

[0103] This embodiment provides a hyperspectral remote sensing image classification device, the device comprising:

[0104] Normalization module: input the original hyperspectral remote sensing image A∈R H×W×B , where H represents the height of the hyperspectral remote sensing image, W represents the width of the hyperspectral remote sensing image, B represents the number of original bands of the hyperspectral remote sensing image, R H×W×B Representing a real number set whose size is H×W×B, and performing normalization processing on the original hyperspectral remote sensing image to obtain a normalized hyperspectral remote sensing image;

[0105] Dimensionality reduction module: used to perform dimensionality reduction processing on the normalized hyperspectral remote sensing image through principal component analysis (PCA) to obtain hyperspectral remote sensing image data A after dimensionality reduction * ∈ R H×W×N , where N represents the number of bands of the hyperspect...

Embodiment 3

[0112] The embodiment of the present invention also provides a hyperspectral remote sensing image classification device, including a processor and a storage medium;

[0113] The storage medium is used to store instructions;

[0114] The processor is configured to operate according to the instructions to perform the steps of the following method:

[0115] Step S1: Input the original hyperspectral remote sensing image A∈R H×W×B , where H represents the height of the hyperspectral remote sensing image, W represents the width of the hyperspectral remote sensing image, B represents the number of original bands of the hyperspectral remote sensing image, R H×W×B Representing a real number set whose size is H×W×B, and performing normalization processing on the original hyperspectral remote sensing image to obtain a normalized hyperspectral remote sensing image;

[0116] Step S2: Perform dimensionality reduction processing on the normalized hyperspectral remote sensing image by princ...

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 remote sensing image classification method and device. The method provided by the invention comprises the following steps: firstly, acquiring rich shallow features by using a convolution layer; then sequentially learning attention weights in spectrum and space through dual-channel attention to assign attention weights to the hyperspectral remote sensing images, amplifying useful information, inhibiting useless information to make the data discrimination degree higher; and finally, taking the weighted 3D data blocks as the input of the dense residual error network to extract spectral spatial features for classification. The attention module fully considers the characteristics of the hyperspectral remote sensing images, learns spectral attention and spatial attention in sequence and assigns attention weights to the hyperspectral remote sensing images in sequence, so that the subsequent dense residual network is facilitated to extract spectral-spatial characteristics with discrimination.

Description

technical field [0001] The invention relates to hyperspectral remote sensing image processing technology, in particular to a hyperspectral remote sensing image classification method and device. Background technique [0002] The hyperspectral remote sensing images captured by the imaging spectrometer, each pixel has hundreds of bands, and the ground objects have obtained a continuous spectral curve with certain characteristics like an "identity label". Spectral remote sensing images can be classified at the pixel level. Hyperspectral remote sensing image classification has a wide range of applications in ecological science, geological science, mineralogy, hydrological science, precision agriculture, and military applications. It has theoretical significance and application value. Hyperspectral remote sensing image classification is mostly based on machine Learning, such as K-means clustering, expectation maximization (EM), support vector machine (SVM), etc. With the develop...

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): G06K9/00G06K9/62G06N3/04
CPCG06V20/194G06V20/13G06N3/045G06F18/213G06F18/214G06F18/241
Inventor 高红民王明霞杨耀曹雪莹李臣明
Owner HOHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com