CAMERA ORIENTATION-INDEPENDENT PARKING EVENTS DETECTION

The paper describes the method for detecting precise position and time of vehicles parking in a parking lot. This task is trivial in case of favorable camera orientation but gets much more complex when an angle between the camera viewing axis and the ground is small. The method utilizes background subtraction and object tracking algorithms for detecting moving objects in a video stream. Objects are classified into vehicles and non-vehicles with SVM based on their shapes. Detection of the time moment when a vehicle parks utilizes spatial and temporal rules. Parking place identification (i.e. where a vehicle has parked) involves distance measurements between the vehicle position and right prisms that represent parking places in 3D space. The paper concludes with results description of a parking event detection based on the real-time analysis of several hours of video streamed from a real parking lot