Check patentability & draft patents in minutes with Patsnap Eureka AI!

Hyperspectral open set domain adaptive method based on multi-classifier domain adversarial network

A multi-classifier and hyperspectral technology, applied in the field of hyperspectral image open set classification, can solve the problems of long time consumption, high cost of data labeling, low classification accuracy, etc., and achieve simple, good classification performance and real-time performance, complexity low effect

Active Publication Date: 2021-08-06
NAT UNIV OF DEFENSE TECH
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a hyperspectral open set field based on a multi-classifier domain confrontation network for the problems of high data labeling cost, time-consuming, and low classification accuracy in the hyperspectral image classification task in the open set scene. adaptive method

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 open set domain adaptive method based on multi-classifier domain adversarial network
  • Hyperspectral open set domain adaptive method based on multi-classifier domain adversarial network
  • Hyperspectral open set domain adaptive method based on multi-classifier domain adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] An embodiment of the present invention provides a hyperspectral open set domain adaptive method based on a multi-classifier domain confrontation network, including:

[0038] S1: Acquire hyperspectral images;

[0039] In this embodiment, the method for acquiring a hyperspectral image includes: providing an original hyperspectral image; performing bilateral filtering preprocessing on the original hyperspectral image to form a hyperspectral image.

[0040] The original hyperspectral image is denoted as P∈H A×B×C , A is the height of the data cube in the original hyperspectral image, B is the width of the data cube in the original hyperspectral image, and C is the number of spectral channels of the data cube in the original hyperspectral image. A hyperspectral image is represented as The spatial dimension size of the data cube in the hyperspectral image is the same as that of the data cube in the original hyperspectral image, and the number of spectral channels of the da...

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

A hyperspectral open set domain adaptive method based on a multi-classifier domain adversarial network evaluates domain-level recognizable feature information of each sample in a target domain based on a dynamic adaptive threshold scheme of a multi-classifier structure, thereby enhancing robustness of an adversarial training process. And finally, the unknown category target is rejected to be an 'unknown' category while the common category of the target domain and the source domain is accurately classified. The method is clear in structure and easy to implement, can obviously improve the hyperspectral image classification effect in an open set scene, and has a deep theoretical basis and practical significance.

Description

technical field [0001] The invention relates to the technical field of hyperspectral image open set classification, in particular to a hyperspectral open set domain adaptive method based on multi-classifier domain confrontation network. Background technique [0002] Due to its rich spatial and spectral information, hyperspectral images are widely used to solve many problems in the field of remote sensing, such as object classification, object segmentation, etc. Among them, object classification is the process of automatically assigning the pixel data in the hyperspectral image to the correct object category label. The early research on object classification is based on manual labeling features, that is, the effective features of the hyperspectral image data are extracted first. , and then use the classifier to classify the features to the correct labels. As deep learning technology has been proved to have strong advantages in data deep feature mining, it has gradually becom...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/084G06V20/194G06V20/13G06F18/22G06F18/214G06F18/24
Inventor 彭元喜唐学斌杨文婧徐炜遐周侗李春潮涂文轩
Owner NAT UNIV OF DEFENSE TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com