Visual Target Detection and Tracking Based on Kalman Filter
In this study, in order to prevent collision and target tracking in autonomous aircraft, Kalman filter with appropriate solutions for target detection and tracking problems are presented. For the Kalman filter application, the motion-based tracking method that facilitates the tracking of multiple objects has been used. A background subtraction algorithm was used for the detection of moving objects while corrective actions were applied to the foreground mask to eliminate noise, and the connected pixel groups to correspond to moving objects were identified. The image used represents an image taken by a hovering drone. Although the tracked multiple targets disappeared behind the obstacle, the estimated location was determined thanks to the Kalman filter. Thanks to the fine tuning of the program codes, a successful follow-up has been achieved. The program has continued to follow the moving shadow as a part of the target.
Copyright (c) 2021 Journal of Aeronautics and Space Technologies
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
The manuscript with title and authors is being submitted for publication in Journal of Aeronautics and Space Technologies. This article or a major portion of it was not published, not accepted and not submitted for publication elsewhere. If accepted for publication, I hereby grant the unlimited and all copyright privileges to Journal of Aeronautics and Space Technologies.
I declare that I am the responsible writer on behalf of all authors.