Detection of Aircraft in Video Images

摘要

NASA is collaborating with the aircraft industry to develop technology for a supersonic passenger airplane called the High-speed Civil Transport (HSCT). One issue that is being examined is the replacement of the conventional forward cockpit windows with synthetic displays. The imagery in these displays would be obtained from video cameras mounted outside the aircraft. A benefit of this configuration is that the video imagery can be examined with computers to determine if another aircraft is in the scene. The goal of this HSCT sub-project is to develop computer vision programs to detect aircraft that are moving in the video images. During FY97, a series of computer programs were written to process video images and to search for moving objects (e.g., other aircraft). Flight tests were conducted in April and May to obtain the video imagery to test the computer programs. Each flight test was conducted with two aircraft. One aircraft was a Boeing 737 with a camera mounted below the nose. With a field-of-view of 13 degrees, this camera recorded the images of a second aircraft (the target plane) flying in various trajectories. For example, one trajectory consisted of the target plane flying from right to left across the path of the Boeing 737, producing the image in the first figure. The position of the target plane i s indicated by the arrow in the figure. In this instance, the target plane was a Beech King Air 200 flying at a distance of 1 nautical mile relatively small in the image.