Ship Detection from Remote Sensing Images Based on Deep Learning

Ziqiang Yuan, Jing Geng*, Tianru Dai

*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Due to the complicated maritime climate environment, the detection of marine Ship by using Remote sensing images is faced with many challenges in the field of object detection. In this paper, a ship detection method based on dark channel priority haze removal and Faster RCNN is proposed to solve this problem. We label and experiment with thousands of ships images on the sea. Compared with the using of object detection model directly and some traditional methods, the detection accuracy of the new method is obviously improved.

Original languageEnglish
Title of host publicationGeo-Spatial Knowledge and Intelligence - 5th International Conference, GSKI 2017, Revised Selected Papers
EditorsFuling Bian, Hanning Yuan, Jing Geng, Chuanlu Liu, Tisinee Surapunt
PublisherSpringer Verlag
Pages336-344
Number of pages9
ISBN (Print)9789811308925
DOIs
Publication statusPublished - 2018
Externally publishedYes
Event5th International Conference on Geo-Spatial Knowledge and Intelligence, GSKI 2017 - Chiang Mai, Thailand
Duration: 8 Dec 201710 Dec 2017

Publication series

NameCommunications in Computer and Information Science
Volume848
ISSN (Print)1865-0929

Conference

Conference5th International Conference on Geo-Spatial Knowledge and Intelligence, GSKI 2017
Country/TerritoryThailand
CityChiang Mai
Period8/12/1710/12/17

Keywords

  • Deep learn
  • Faster RCNN
  • Haze removal
  • Remote sensing

Fingerprint

Dive into the research topics of 'Ship Detection from Remote Sensing Images Based on Deep Learning'. Together they form a unique fingerprint.

Cite this