Unlock instant, AI-driven research and patent intelligence for your innovation.

High-resolution remote sensing scene classification method based on multi-scale CENTRIST characteristics

A scene classification, multi-scale technology, applied in scene recognition, instrument, character and pattern recognition, etc., can solve the problems of high classification accuracy and less than ideal classification accuracy, and achieve the effect of good discrimination and robustness

Inactive Publication Date: 2018-11-13
HOHAI UNIV
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Although the above-mentioned existing technologies have their own characteristics, they generally have high classificatio

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-resolution remote sensing scene classification method based on multi-scale CENTRIST characteristics
  • High-resolution remote sensing scene classification method based on multi-scale CENTRIST characteristics
  • High-resolution remote sensing scene classification method based on multi-scale CENTRIST characteristics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail:

[0036] Although CENTRIST can extract the texture features of the image, because its neighborhood size is only 3×3, it is not enough to fully express the texture features of a larger area. How to effectively describe the texture features of a larger area and how to integrate multi-scale information is a problem Questions to consider.

[0037] image 3 It is a flow chart of remote sensing scene classification based on multi-scale CENTRIST features. Based on the above ideas, the present invention first divides remote sensing images into several sub-images by using spatial pyramids, and then uses multi-scale CENTRIST for feature extraction and concatenation of each sub-image. It is represented as multi-scale CENTRIST features, and finally the multi-scale CENTRIST features of different sub-images are concatenated to form the overall feature of the imag...

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-resolution remote sensing scene classification method based on multi-scale CENTRIST characteristics. The method comprises the steps of firstly decomposing an image intosub-images according to a spatial pyramid structure; then for the sub-images, extracting the CENTRIST characteristics of all scales by adopting a multi-scale CENTRIST operator; fusing the CENTRIST characteristics of all the scales; serially connecting the multi-scale CENTRIST characteristics of the sub-images to form global characteristic expression of a scene; and finally performing classification by adopting an SVM algorithm. According to the method, multi-scale expression of image texture information can be realized, so that the classification precision is improved.

Description

technical field [0001] The invention relates to the technical field of remote sensing scene classification, in particular to a remote sensing scene classification method based on multi-scale CENTRIST features. Background technique [0002] With the rapid development of remote sensing technology, the number of high-resolution remote sensing images has also increased dramatically, and the demand for scene classification of high-resolution remote sensing images has become more and more urgent. Compared with low- and medium-resolution remote sensing images, high-resolution images can provide clearer spatial layout of ground objects, finer texture and size information, and richer geometric structures, but at the same time bring spectral differences of similar ground objects enhancement, and the reduction of spectral heterogeneity of different ground objects. Therefore, due to the complexity and variability of remote sensing scenes, intelligent scene classification becomes a chal...

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): G06K9/00G06K9/46G06K9/62
CPCG06V20/13G06V10/44G06V10/464G06F18/2411
Inventor 袁宝华韩立新勾智楠
Owner HOHAI UNIV