Event-driven traffic ticketing system
This thesis presents a new schema to implement an event-driven traffic ticketing system. The entire system consists of four modules, namely, (1) event detection module, (2) plate number recognition module to recognize the extracted plate number, (3) database management module to execute information retrieval and (4) traffic ticket transmission module to deliver traffic tickets. Due to the importance of plate number recognition module in entire system, this thesis focuses on plate number localization and recognition. A novel plate number localization algorithm called Secondary Positioning (SP) method is proposed in this thesis. The rough position of plate number is identified at first stage by searching the red light region in HSV colour space, and then the accurate position of plate number is localized in the second stage by finding out the vertical edge of plate number. Template matching implemented for recognizing individual plate character incorporates correction coefficient calculated between templates and testing images. The test datasets include two subsets, namely, (1) 120 images for plate number localization and (2) 80 images for plate number recognition. The success rates obtained from plate number localization and recognition are approximately 75% and 70% respectively.