High spectral classification result optimization method combining with space information

A hyperspectral classification and optimization method technology, applied in the field of hyperspectral classification result optimization combined with spatial information, can solve problems such as ignoring information and poor results

Active Publication Date: 2014-06-25
奥谱天成(湖南)信息科技有限公司
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When traditional high-level spectral remote sensing images are classified and recognized, they often only focus on the information in the spectral dimension of the data, but ignore the information contained in the spatial dimension, and the general effect is not good.

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
  • High spectral classification result optimization method combining with space information
  • High spectral classification result optimization method combining with space information
  • High spectral classification result optimization method combining with space information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] According to the spatial domain information, the present invention first proposes a method for obtaining the edge profile of a hyperspectral image with an adaptive threshold: the traditional method of edge extraction using single-band data and a single threshold is not effective, so the present invention first uses multiple The gradient map is extracted from the data of each band, and then a set of adaptive thresholds are used for edge extraction.

[0049] Then an internal expansion method was proposed. After the edge profile is obtained, a completely closed edge profile is required for optimization using the traditional region growing method, but this is difficult to obtain in practical applications, and the internal dilation method proposed by the present invention can be used even for incompletely closed edge profiles. can achieve better results.

[0050] Combining the two methods can optimize the hyperspectral classification results based on SVM support vector mach...

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 high spectral classification result optimization method combining with space information. A conventional high spectral image classification technology mainly focuses on how to use the classification information of spectral spaces better, but ignores the information of image spatial domains. According to the invention, in the process of carrying out self spectral feature classification by using self data, spectral classification results are supplemented by adopting spatial domain effective information combining self-adaptive threshold edge extraction with an internal expansion method. According to the invention, an SVM (support vector machine) based classification method is adopted for carrying out spectral domain classification on data firstly; then, effective spatial domain information is introduced by using the self-adaptive threshold edge and the internal expansion method so as to correct spectral classification results. According to the invention, information contained in hyperspectral data is used more fully, and the precision of hyperspectral image classification is improved.

Description

technical field [0001] The invention belongs to the field of information technology, relates to pattern recognition and image processing technology, in particular to a hyperspectral classification result optimization method combined with spatial information. Background technique [0002] Hyperspectral remote sensing data has high spectral resolution and contains rich spectral information of surface features. When traditional high-level spectral remote sensing images are classified and recognized, they often only focus on the information in the spectral dimension of the data, but ignore the information contained in the spatial dimension, and the general effect is not good. In fact, hyperspectral remote sensing images can express ground objects from two different perspectives: spatial dimension and spectral dimension. When performing spectral dimension analysis on hyperspectral data, the introduction of spatial dimension information can increase a large amount of hidden infor...

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
Inventor 郭宝峰陈春种吴香伟彭冬亮谷雨左燕
Owner 奥谱天成(湖南)信息科技有限公司
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