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

Robust classification method for hyperspectral images based on segmented depth features and low-rank representation

A hyperspectral image and depth feature technology, which is applied in the fields of instrumentation, computing, character and pattern recognition, etc., can solve the problems of not fully exploiting the spatial correlation of hyperspectral images, limiting the classification accuracy of hyperspectral images, etc., and achieve good classification results Effect

Active Publication Date: 2021-09-07
NORTHWESTERN POLYTECHNICAL UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method only considers the inter-spectral correlation, and does not fully exploit the spatial correlation of hyperspectral images, as well as intra-class similarity and inter-class difference, thus limiting the accuracy of hyperspectral image classification

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
  • Robust classification method for hyperspectral images based on segmented depth features and low-rank representation
  • Robust classification method for hyperspectral images based on segmented depth features and low-rank representation
  • Robust classification method for hyperspectral images based on segmented depth features and low-rank representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031]The present invention proposes a hyperspectral image robust classification method based on segmented depth features and low-rank representation, the specific process is as follows:

[0032] 1. Calculate the correlation coefficient between spectra

[0033] Hyperspectral images include hundreds of continuous spectral bands, and because of this continuity, there is a strong correlation between the bands. In order to better explore the correlation of different spectral regions, the correlation coefficient between different band spectra of hyperspectral images is calculated, and then the connected spectral bands with correlation coefficients greater than 0 are divided into one section, and then the original hyperspectral image is divided into spectral dimensions segment.

[0034] 2. Training stack denoising autoencoder

[0035] First, the hyperspectral image pixels are randomly divided into training data and test data, and according to the spectral dimension division method...

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 provides a hyperspectral image robust classification method based on segmented depth features and low-rank representation. First, in order to reduce the impact of noise on feature extraction as much as possible, unsupervised feature extraction is performed on hyperspectral images using a stack-based denoising autoencoder network; then, by fully mining the similarity and inter-class A robust classifier based on low-rank representation is established; finally, an effective optimization method is used to optimize and solve the objective function. In the case of noise in the training data, better classification results can also be obtained.

Description

technical field [0001] The invention belongs to the technical field of hyperspectral image processing, and in particular relates to a hyperspectral image robust classification method based on segmented depth features and low-rank representation. Background technique [0002] At present, hyperspectral images have been widely used in the fields of resource exploration, environmental monitoring and target recognition, but in practical applications, due to the influence of imaging environment and transmission process, hyperspectral images are susceptible to noise interference, resulting in image quality degradation, thus Affects the accuracy of hyperspectral image interpretation. How to effectively complete the hyperspectral image classification task in the presence of widespread noise has gradually attracted widespread attention from scholars at home and abroad. A large number of experiments have shown that stacked autoencoders are an effective unsupervised feature learning me...

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 Patents(China)
IPC IPC(8): G06K9/62G06K9/46
CPCG06V10/40G06V10/58G06F18/28G06F18/2155G06F18/24
Inventor 魏巍张艳宁王聪张磊
Owner NORTHWESTERN POLYTECHNICAL 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