Factors that influence the geometric detection pattern of vehicle-based licence plate recognition camera systems

Abstract

Licence plate recognition (LPR) systems are used to automatically extract the characters from licence plates positioned in front of a camera. The geometric detection pattern is the region within which the system can accurately recognise licence plates and is of special interest when the system is mounted in a moving vehicle. In this research, the theory surrounding camera optics was investigated and used as the basis of a software simulation model. Inspired by the simulation measurements, a real-world experimental test was conducted to further explore the influence various factors have on the geometric detection pattern. Analysis of these measurements provided greater insight how multiple factors individually contribute to the shape and size of the geometric detection pattern and serves as a guide in the design of vehicle-based LPR systems