An Automated System for Performance Assessment of Airport Lighting

摘要

This paper presents work undertaken into the development of an automated air-based vision system for assessing the performance of an approach lighting system (ALS) installation in accordance with International Civil Aviation Organisation (ICAO) standards. The measuring device consists of an image sensor with associated lens system fitted to the interior of an aircraft. The vision system is capable of capturing sequences of airport lighting images during a normal approach to the airport. These images are then processed to determine the uniformity of the ALS. To assess the uniformity of the ALS the luminaires must first be uniquely identified and tracked through an image sequence. A model-based matching technique is utilised which uses a camera projection system to match a set of template data to the extracted image data. From the matching results the associated position and pose of the camera is estimated. Each luminaire emits an intensity which is dependant on its angular displacement from the camera. As such, it is possible to predict the intensity that each luminaire within the ALS emits during an approach. Luminaires emitting the same intensity are banded together for the uniformity analysis. Uniformity assumes that luminaires in close proximity exhibit similar luminous intensity characteristics. During a typical approach grouping information is obtained for the various sectors of luminaires. This grouping information is used to compare luminaires against one another in terms of their extracted grey level information. The developed software is validated using data acquired during an actual approach to a UK airport.