Share Email Print

Optical Engineering

Detection of aircraft in video sequences using a predictive optical flow algorithm
Author(s): Jeffrey W. McCandless
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This paper presents a computer vision algorithm that segregates spurious optical flow artifacts to detect a moving object. The algorithm consists of six steps. First, the pixels in each image are shifted to compensate for camera rotation. Second, the images are smoothed with a spatiotemporal Gaussian filter. Third, the optical flow is computed with a gradient-based technique. Fourth, optical flow vectors with small magnitudes are discarded. Fifth, vectors with similar locations, magnitudes, and directions are clustered together using a spatial consistency test. Sixth, similar optical flow vectors are extended temporally to make predictions about future optical flow locations, magnitudes, and directions in subsequent frames. The actual optical flow vectors that are consistent with those predictions are associated with a moving object. This algorithm was tested on images obtained with a video camera mounted below the nose of a Boeing 737. The camera recorded two sequences containing a second flying aircraft. The algorithm detected the aircraft in 82% of the frames from the first sequence and 78% of the frames from the second sequence. In each sequence, the false-alarm rate was zero. These results illustrate the effectiveness of using a comprehensive predictive technique when detecting moving objects.

Paper Details

Date Published: 1 March 1999
PDF: 8 pages
Opt. Eng. 38(3) doi: 10.1117/1.602130
Published in: Optical Engineering Volume 38, Issue 3
Show Author Affiliations
Jeffrey W. McCandless, NASA Ames Research Ctr. (United States)

© SPIE. Terms of Use
Back to Top