Pedestrian Detection for Driver Assistance Using
Massimo Bertozzi, Alberto Broggi, A. Fascioli, Thorsten Graf, Marc-Michael Meinecke · 2004
This paper describes a system for pedestrian detec- tion in infrared images, which has been implemented on an exper- imental vehicle equipped with an infrared camera. The proposed system has been tested in many situations and has proven to be ef- ficient and with a very low false-positive rate. It is based on a mul- tiresolution localization of warm symmetrical objects with specific size and aspect ratio; anyway, because road infrastructures and other road participants may also have such characteristics, a set of matched filters is included in order to reduce false detections. A final validation process, based on human shape's morphological characteristics, is used to build the list of pedestrian appearing in the scene. Neither temporal correlation nor motion cues are used in this first part of the project: the processing is based on the anal- ysis of single frames only.