Cloud-Resilient Aircraft Tracking: An Antidrift Multifilter Approach For Remote Sensing video

  • Unique Paper ID: 167513
  • Volume: 11
  • Issue: 3
  • PageNo: 1558-1565
  • Abstract:
  • Aircraft tracking in satellite video data holds paramount importance in various domains such as military operations, airport management, and aircraft rescue missions. This paper introduces an innovative approach, combining correlation and Kalman filtering, to develop an antidrift multifilter tracker tailored for this purpose. We propose a novel temporal consistency-constrained background-aware correlation filter algorithm, integrating temporal regularization to combat model drift caused by cloud occlusion, thereby enhancing tracking accuracy. Our experimental evaluations demonstrate superior antidrift performance compared to contemporary methods, particularly in scenarios involving cloud occlusion, while maintaining stability in complex conditions. Additionally, we present an extension by incorporating diverse techniques including ADMFT and YOLO variants (v5, v6, v7, v8) for dataset analysis. Moreover, to facilitate user testing and validation, we propose integrating this solution into a frontend utilizing the Flask framework with authentication features. We anticipate that our model will offer valuable insights for researchers interested in satellite video object tracking, especially in mitigating challenges posed by cloud occlusion.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 3
  • PageNo: 1558-1565

Cloud-Resilient Aircraft Tracking: An Antidrift Multifilter Approach For Remote Sensing video

Related Articles