Skip to main navigation Skip to search Skip to main content

Maritime Target Saliency Detection for UAV Based on Raptor Vision Perception Enhancement Mechanism

  • Beihang University

Research output: Contribution to journalArticlepeer-review

Abstract

Inspired by the enhanced perception mechanism of raptor vision in various hunting environments, we propose a biological raptor vision perception enhancement algorithm for unmanned aerial vehicles (UAVs) to detect medium-distance maritime targets in complex and dynamic ocean environments. First, we summarize the various perception enhancement capabilities based on the physiological structure and functional characteristics of the raptor visual system. Second, we establish mathematical models for light perception enhancement, foreground perception enhancement, contrast perception enhancement, and perception balance based on the raptor vision's enhanced perception mechanism. Finally, our proposed biological raptor vision perception enhancement method is experimentally tested in different scenarios and compared with seven other algorithms. The results of the comparative experiments and relevant evaluation indicators show that the algorithm effectively retains the feature information of maritime targets while suppressing interference from the environmental background. This maritime target saliency detection algorithm can be applied to the vision system of sea UAVs, providing a feasible solution for the medium-distance recovery of UAVs in complex and dynamic ocean environments.

Original languageEnglish
Pages (from-to)70-83
Number of pages14
JournalIEEE Aerospace and Electronic Systems Magazine
Volume41
Issue number5
DOIs
StatePublished - 1 May 2026

Keywords

  • perception enhancement mechanism
  • raptor biological vision
  • target saliency detection
  • unmanned aerial vehicles

Fingerprint

Dive into the research topics of 'Maritime Target Saliency Detection for UAV Based on Raptor Vision Perception Enhancement Mechanism'. Together they form a unique fingerprint.

Cite this