LR-SARNET: A Lightweight and Robust Network for Multi-scale and Multi-scene SAR Ship Detection

Shibo Chang, Xiongjun Fu*, Jian Dong, Hao Chang

*Corresponding author for this work

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

Abstract

SAR Ship image has the characteristics of complex background, blurred target edge and scale difference, which makes the target detection difficult. This article builds a lightweight and robust network for multi-scale and multi-scene SAR ship detection (LR-SARNET). Firstly, with the goal to minimize the computational complexity of the model, a lightweight backbone feature extraction network (CGNet) is designed to generate sufficiently redundant feature maps with low computational cost. Secondly, a linear feature fusion module (ENECK) is designed to efficiently fuse deep local feature maps. Finally, the extremely efficient spatial pyramid (EESP) is integrated into the target detection head, which expands the receptive field of the network. The experiment on SSDD and HRSID dataset proves that our algorithm has strong robustness and excellent generalization performance.

Original languageEnglish
Title of host publicationImage and Graphics Technologies and Applications - 18th Chinese Conference, IGTA 2023, Revised Selected Papers
EditorsWang Yongtian, Wu Lifang
PublisherSpringer Science and Business Media Deutschland GmbH
Pages456-471
Number of pages16
ISBN (Print)9789819975488
DOIs
Publication statusPublished - 2023
Event18th Chinese Conference on Image and Graphics Technology and Application Conference, IGTA 2023 - Beijing, China
Duration: 17 Aug 202319 Aug 2023

Publication series

NameCommunications in Computer and Information Science
Volume1910 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference18th Chinese Conference on Image and Graphics Technology and Application Conference, IGTA 2023
Country/TerritoryChina
CityBeijing
Period17/08/2319/08/23

Keywords

  • Anchor-free mechanism
  • Intensive object detection
  • SAR image
  • Ship detection
  • Small target detection

Fingerprint

Dive into the research topics of 'LR-SARNET: A Lightweight and Robust Network for Multi-scale and Multi-scene SAR Ship Detection'. Together they form a unique fingerprint.

Cite this