The project is part of a combined project designed to develop a real time License Plate Recognition system.
Abstract
The project is part of a combined project designed to develop a real time License Plate Recognition system. The system gives a report of each and every vehicles number captured in the video camera attached to the system. Optional uses of this system are to: control the gate of a private parking lot; give a report of the vehicles passing through a certain road; spot stolen or wanted cars, etc.
This part implements the first part of the system: analyzing the video movie and locating the License Plate in a video frame.
Figure 1 – This is the final system graphical user interface. The upper left window is the preview window constantly showing the real time image that is captured in the video camera. The lower left window is the captured window showing the frame that the system is currently analyzing. The numbers window shows all the numbers of the License Plates that were recognized so far.
Acknowledgments
We would like to thank our instructor and lab chief engineer, Johanan Erez, for his help and support.
We are also grateful to the Ollendorf Minerva Center Fund for their contribution to the Vision and Imaging Science Laboratory.
The Algorithms
In order to solve the problem, we came up with two algorithms, based on basic methods in image processing:
The algorithm of locating a License Plate in a still picture and the algorithm of selecting frames most likely to contain a License Plate.
The algorithm of selecting frames most likely to contain a License Plate
The algorithm of locating a License Plate in a still picture
