Identification of License Plate Number of Vehicle from Video: Mathematical Morphology, Projection & OCR Techniques

Abstract

During past few years, as number of vehicles increased on the roads, managing traffics has been a big challenge for government. To catch and punish traffic offender, lots of man power is required. The automatic surveillance system can make this job easier. As the number plate is meaning to identify the vehicle, Automatic license Plate recognition, (ALPR) is substantial part to make system automated. ALPR is a system in which, every vehicle passing through the view of camera are monitored. Automated License Plate Reading Systems have been used in transportation applications such as parking lot management; destination studies/trip surveys, cordon studies, travel time studies and automated traffic ticketing systems. As this project is for Indian vehicles, the challenge in this project assumes all vehicles following the Indian traffic rules. The thesis is divided into two modules: Finding the Number plate from the Frame and character extraction and recognition. Here the number plate extraction is implemented based on the property that there are more horizontal and vertical edges nearer number plate region. The first module implements mathematical morphology in conjunction with projection method, which finds and segments the number plate region. This method gives promising results. For character extraction and recognition template matching and statistical methods are used.