Pedestrian detection in infrared images

Abstract — This paper describes an approach for pedestrian detection in infrared images. The developed system has been implemented on an experimental vehicle equipped with an infrared camera and preliminarily tested in different situations. It is based on the localization of warm symmetrical objects with specific size and aspect ratio; since also road infrastructures and other road participants may have such characteristics, a set of matched filters was added in order to reduce false detections. A final validation process, based on the human shape’s morphological characteristics, is used to build the list of pedestrian appearing in the scene. No temporal correlation, nor motion cues are used in this first part of the project. Keywords — Pedestrian Detection, Machine Vision, IR Imagery. I