Change detection in multitemporal synthetic aperture radar images using dual-channel convolutional neural network

Tao Liu, Ying Li, Ying Cao, Qiang Shen

Research output: Contribution to journalArticlepeer-review

31 Citations (Scopus)
346 Downloads (Pure)

Abstract

This paper proposes a model of dual-channel convolutional neural network (CNN) that is designed for change detection in SAR images, in an effort to acquire higher detection accuracy and lower misclassification rate. This network model contains two parallel CNN channels, which can extract deep features from two multitemporal SAR images. For comparison and validation, the proposed method is tested along with other change detection algorithms on both simulated SAR images and real-world SAR images captured by different sensors. The experimental results demonstrate that the presented method outperforms the state-of-the-art techniques by a considerable margin
Original languageEnglish
Article number042615
JournalJournal of Applied Remote Sensing
Volume11
Issue number4
DOIs
Publication statusPublished - 12 Oct 2017

Keywords

  • SAR image
  • change detection
  • deep learning
  • dual-channel convolutional neural network

Fingerprint

Dive into the research topics of 'Change detection in multitemporal synthetic aperture radar images using dual-channel convolutional neural network'. Together they form a unique fingerprint.

Cite this