Intensity and edge-based symmetry detection with an application to car-following

摘要: Abstract We present two methods for detecting mirror symmetry in images, one based directly on the intensity values and another a discrete representation of local orientation. A finder has been developed which uses intensity-based method to search an image compact regions display some degree due similarities across straight axis. In different approach, we look at as bilateral relationship between orientations. symmetry-enhancing edge detector is presented indicates edges dependent orientations positions. SEED, call it, element implemented by feedforward network that holds conditions. use SEED find contours symmetric objects know axis from finder. The described this paper have tested recognition tracking cars real-time system automatic car-following headway control normal roads.