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

Hyperspectral image characteristic parallel extraction and classification method

A technology of hyperspectral image and classification method, which is applied in the field of parallel extraction and classification of hyperspectral image features, and can solve the problems of strong redundancy and large feature dimension.

Inactive Publication Date: 2016-10-12
HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
View PDF0 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the current three-dimensional Gabor features have good performance, there are also shortcomings such as large feature dimensions and strong redundancy.

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 characteristic parallel extraction and classification method
  • Hyperspectral image characteristic parallel extraction and classification method
  • Hyperspectral image characteristic parallel extraction and classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0037] The present invention mainly studies how to extract space-spectral domain joint change features from hyperspectral images, and filter and fuse these features through intelligent optimization algorithms, remove redundant features, and finally achieve accurate and efficient recognition and classification effects. The present invention focuses on the joint change feature extraction based on the three-dimensional Gabor wavelet in the space-spectral domain, specifically as follows:

[0038] (1) Parallel feature extraction based on 3D Gabor wavelet joint change in space-spectral domain

[0039] The present invention optimizes these coding methods while deeply studying the mechanism of simultaneously extracting signal variation characteristics in the three-dimensional space domain and the frequency domain, and focuses on the coding method that ...

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 characteristic parallel extraction and classification method. By adopting three-dimensional spatial domain and frequency domain combination analysis, parallel characteristic extraction of three-dimensional and hyperspectral images is carried out, and then the characteristics are enhanced and integrated, and in addition, influences of noises on accuracy are greatly reduced by fully using rich information of high-dimensional data and a structural relationship between the characteristics. Various phases, directions, frequency domains, and three-dimensional space coding methods are deeply researched, and by starting from structural relationships between directions and frequencies of various Gabor characteristics, changing characteristics of signals of a spatial-spectral domain and the frequency domain are extracted in a combined way, and at the same time, an intelligent algorithm is innovatively introduced in waveband selection, and therefore under a precondition of guaranteeing identification accuracy, redundant information is reduced, identification efficiency is improved, and a wide application and popularization prospect is provided.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method for parallel extraction and classification of hyperspectral image features. Background technique [0002] Hyperspectral imaging technology is an image data technology based on very narrow bands developed in the past two decades. Its most prominent application is in the field of remote sensing detection, and it has broad application prospects in more and more civilian fields. Hyperspectral imaging mainly takes pictures of objects through light sources of different wavelengths, and obtains photos of objects in different wavelength bands, thereby extracting richer useful information. Color imaging obtains multispectral images through three channels of red, blue, and green, while hyperspectral technology usually divides visible light and near-infrared into hundreds of bands and hundreds of hyperspectral images. The spectral resolution of hyperspectral r...

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): G06K9/62G06K9/46
CPCG06V10/40G06V10/58G06F18/22G06F18/2113G06F18/2411G06F18/251
Inventor 王岢张海军李旭涛叶允明徐晓飞
Owner HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
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