Automatic aircraft extraction using video matting and frame registration

• Published in: IET image processing Vol. 14; no. 8; pp. 1628 - 1635
• Main Authors: Amin, Benish, Mohsin Riaz, Muhammad, Ghafoor, Abdul
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 19.06.2020
• Subjects: alpha mattes; automatic aircraft extraction; computational complexity; different videos; feature extraction; final alpha matte; given ROI; high quality alpha maps; image colour analysis; image processing; image registration; image segmentation; image sequences; KNN matting algorithm; medical image processing; nonrigid image registration; object detection; object tracking; object tracking algorithm; Research Article; static region; video matting; video sequence; video signal processing; image processing; video matting; object tracking algorithm; different videos; nonrigid image registration; alpha mattes; image registration; final alpha matte; video sequence; high quality alpha maps; object detection; given ROI; automatic aircraft extraction; feature extraction; image segmentation; KNN matting algorithm; object tracking; static region; image colour analysis; video signal processing; image sequences; medical image processing; computational complexity
• ISSN: 1751-9659; 1751-9667; 1751-9667
• DOI: 10.1049/iet-ipr.2019.0067

This study presents a technique to automatically extract the aircraft from video sequences. The technique incorporates object tracking algorithm to separate the region of interest (ROI), i.e. movement of aircraft, from video sequence, to reduce the computational complexity. Trimaps are generated automatically for given ROI using non-rigid image registration to obtain the prior information of the object and alpha mattes are estimated for each frame using KNN matting algorithm. However, the alpha matte of static region is estimated only once, that is combined with the alpha matte of moving region to generate final alpha matte. Simulations performed on different videos show that the proposed technique not only reduces the computational complexity, but also gives competitive results by generating high quality alpha maps.