Small objects tracking in satellite videos based on feature enhancement and motion estimation

Cunbao Lin; Zhaoming Li; Xinyan Yang; Shuhan Du; Haining Wang

doi:10.1117/12.2691632

6 November 2023 Small objects tracking in satellite videos based on feature enhancement and motion estimation

Cunbao Lin, Zhaoming Li, Xinyan Yang, Shuhan Du, Haining Wang

Proceedings Volume 12921, Third International Computing Imaging Conference (CITA 2023); 129213N (2023) https://doi.org/10.1117/12.2691632
Event: Third International Computing Imaging Conference (CITA 2023), 2023, Sydney, Australia

Abstract

Small objects tracking in satellite videos faces challenge due to their small size and object occlusion. A novel correlation filter-based tracker combined with feature enhancement and motion estimation is proposed. The guided filtering method is introduced to realize the edge-preserving detail enhancement. Meanwhile, to fully exploit the advantage of the edge-preserving enhanced images, the Laplacian of Gaussian (LOG) edge feature and the Histogram of Oriented Gradient (HOG) textual feature are simultaneously employed to obtain the fused response. Besides, since small objects in satellite videos are susceptible to be occluded, an adaptive Kalman filtering is designed to estimate the motion of the targets, and a decision-level fusion strategy is developed for the final trajectory determination of the small objects. The proposed tracker is evaluated on the SatSOT dataset compared with 9 state-of-the-art trackers. The results reveal that the precision of 73.67% and the success rate of 41.83% are respectively obtained, which demonstrated the effectiveness of the proposed method.

Citation Download Citation

Cunbao Lin, Zhaoming Li, Xinyan Yang, Shuhan Du, and Haining Wang "Small objects tracking in satellite videos based on feature enhancement and motion estimation", Proc. SPIE 12921, Third International Computing Imaging Conference (CITA 2023), 129213N (6 November 2023); https://doi.org/10.1117/12.2691632

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available