Parking lot monitoring system using an autonomous quadrotor UAV

This work aims to investigate the use of a drone-based system to recognize license plates of vehicles in a parking lot. Many parking lots contain surveillance cameras mounted on walls or light towers for indoor and outdoor lots respectively. Parked vehicles are commonly monitored by law enforcement agents by driving around the parking lot to identify license plates with an onboard camera or a handheld device. However, these systems use expensive hardware and proprietary software. The main goal of this thesis is to develop an autonomous parking lot surveillance system using low-cost hardware and open-source software. Similar to wall-mounted surveillance cameras, a drone-based system can monitor parking lots without affecting the flow of traffic while also offering the mobility of patrol vehicles. The Parrot AR Drone 2.0 is the quadrotor drone used in this work due to its modularity and cost efficiency. Video and navigation data (including GPS) are communicated to a host computer using a Wi-Fi connection. The host computer analyzes navigation data using a custom flight control loop to determine control commands to be sent to the drone. A new license plate recognition pipeline is used to identify license plates of vehicles from inflight video footage. Although license plate recognition is a well studied topic, previous academic works have exploited discernibility between characters in high resolution images obtained from stationary cameras. In this work, the motion of the camera presents a novel challenge to the task of license plate recognition.