Title:
HYPERSPECTRAL IMAGE BAND SELECTION METHOD AND SYSTEM BASED ON LATENT FEATURE FUSION
Document Type and Number:
WIPO Patent Application WO/2022/227914
Kind Code:
A1
Abstract:
Disclosed in the present application are a hyperspectral image band selection method and system based on latent feature fusion. The hyperspectral image band selection method based on latent feature fusion comprises: S11, inputting a hyperspectral image cube, and segmenting the input hyperspectral image cube into several areas by using super-pixel segmentation; S12, respectively learning, from the several areas, low-dimensional latent features corresponding to the several areas, so as to obtain latent feature matrices of all areas; S13, calculating an average Laplacian matrix and an average latent feature matrix of the hyperspectral image cube; S14, fusing the latent feature matrices of all the areas, the average Laplacian matrix and the average latent feature matrix to obtain a low-dimensional self-representation matrix of the hyperspectral image cube; and S15, clustering the obtained low-dimensional self-representation matrix by using a K-means algorithm, so as to obtain an optimal band subset of the hyperspectral image cube.
More Like This:
Inventors:
ZHU XINZHONG (CN)
XU HUIYING (CN)
TANG CHANG (CN)
ZHAO JIANMIN (CN)
XU HUIYING (CN)
TANG CHANG (CN)
ZHAO JIANMIN (CN)
Application Number:
PCT/CN2022/081429
Publication Date:
November 03, 2022
Filing Date:
March 17, 2022
Export Citation:
Assignee:
UNIV ZHEJIANG NORMAL (CN)
International Classes:
G06K9/62
Foreign References:
CN113269201A | 2021-08-17 | |||
CN105184314A | 2015-12-23 | |||
CN103914705A | 2014-07-09 | |||
CN106529508A | 2017-03-22 | |||
CN105989592A | 2016-10-05 |
Attorney, Agent or Firm:
BEIJING TIANDA INTELLECTUAL PROPERTY OFFICE (CN)
Download PDF: