Skip to main navigation Skip to search Skip to main content

Change detection method for hyperspectral image with sequential time series inputs

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a novel change detection method of hyperspectral data based on convolutional feature extraction and long short term memory unit. To extract abstract nonlinear features of the input data, we choose deepleaming-based method, Convolutional Neural Network, to avoid the drawbacks of linear methods. Data from different period of time are processed separately. A flatten layer helps the extracted feature maps to reconstruct as a new feature vector. Experiments show the effectiveness of the proposed model compared with traditional change detection methods. The proposed method performs well in both changed samples and unchanged samples especially for hyperspectral images.

Original languageEnglish
Title of host publicationProceedings of the 36th Chinese Control Conference, CCC 2017
EditorsTao Liu, Qianchuan Zhao
PublisherIEEE Computer Society
Pages10819-10822
Number of pages4
ISBN (Electronic)9789881563934
DOIs
StatePublished - 7 Sep 2017
Event36th Chinese Control Conference, CCC 2017 - Dalian, China
Duration: 26 Jul 201728 Jul 2017

Publication series

NameChinese Control Conference, CCC
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference36th Chinese Control Conference, CCC 2017
Country/TerritoryChina
CityDalian
Period26/07/1728/07/17

Keywords

  • change detection
  • convolutional neural network
  • long short term memory unit
  • sequential time series

Fingerprint

Dive into the research topics of 'Change detection method for hyperspectral image with sequential time series inputs'. Together they form a unique fingerprint.

Cite this