Title:
NEURAL NETWORK OPTIMIZATION METHOD FOR REMOTE SENSING IMAGE CLASSIFICATION, AND TERMINAL AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2022/052367
Kind Code:
A1
Abstract:
The present application relates to a neural network optimization method for remote sensing image classification, and a terminal and a storage medium. The method comprises: acquiring a remote sensing image data set; constructing an anti-noise network model, wherein the anti-noise network model comprises an image segmentation model and a loss selection model, and the image segmentation model is a U-Net network based on an SE module; and inputting the remote sensing image data set into the anti-noise network model for iterative training, performing, via the anti-noise network model, image segmentation by means of the U-Net network based on the SE module, so as to obtain an image classification result, using, via the loss selection model, a ksigma criterion to perform loss selection, and removing an error that exceeds a set deviation interval, so as to obtain an optimal network model parameter. By means of the embodiments of the present application, the feature extraction capability of a network model is improved, and the problem of a decrease in the classification precision of a neural network caused by noise of tags in a remote sensing image data set is solved.
Inventors:
LIN CHUANG (CN)
CHEN JINSONG (CN)
LI HONGZHONG (CN)
CHEN JINSONG (CN)
LI HONGZHONG (CN)
Application Number:
PCT/CN2020/138818
Publication Date:
March 17, 2022
Filing Date:
December 24, 2020
Export Citation:
Assignee:
SHENZHEN INST OF ADV TECH CAS (CN)
International Classes:
G06K9/62
Foreign References:
| CN110211137A | 2019-09-06 | |||
| CN110443143A | 2019-11-12 | |||
| US20190057245A1 | 2019-02-21 |
Other References:
LI WANQI, LI KEJIAN;CHEN SHAOBO: "Multi-modal fusion based method for high resolution remote sensing image segmentation", ZHONGNAN MINZU DAXUE XUEBAO (ZIRAN KEXUE BAN) -
JOURNAL OF SOUTH-CENTRAL UNIVERSITY FOR NATIONALITIES ( NATURAL SCIENCE EDITION, ZHONGNAN MINZU DAXUE, CN, vol. 39, no. 4, 31 August 2020 (2020-08-31), CN , pages 405 - 412, XP055911374, ISSN: 1672-4321, DOI: 10.12130/znmdzk.20200412
LIU HAO, JIANCHENG LUO, BO HUANG, HAIPING YANG, XIAODONG HU, NAN XU, LIEGANG XIA: "Building Extraction based on SE-Unet", JOURNAL OF GEO-INFORMATION SCIENCE, vol. 21, no. 11, 25 November 2019 (2019-11-25), pages 1779 - 1789, XP055911485, ISSN: 1560-8999
LIU HAO, JIANCHENG LUO, BO HUANG, HAIPING YANG, XIAODONG HU, NAN XU, LIEGANG XIA: "Building Extraction based on SE-Unet", JOURNAL OF GEO-INFORMATION SCIENCE, vol. 21, no. 11, 25 November 2019 (2019-11-25), pages 1779 - 1789, XP055911485, ISSN: 1560-8999
Attorney, Agent or Firm:
BEIJING ZHONG XUN TONG DA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: