A Systematic Performance Comparison of Artificial Intelligence Techniques used for ALNPR System

For transport planning and engineering systems, the Automated Licensed Number plate Recognition (ALNPR) can provide a valuable data source. From the different survey points the multiple tasks can be handled by ALNPR systems. The captured image of the vehicle licensed number plate, the registration number plate location and recognition can be analyzed by the Automated Licensed Number Plates based system. The parameters of the ALNPR systems are affected by the different problems observed while implementing this system by using techniques which are used for number plate detection and character recognition. The detection from the deviation between crossing vehicle and original vehicle number plate was examined. The detection rate cannot be emphasized because it depends on the variable factors like intensity and angle of the sun, low illumination situations, speed of vehicle, deformed or dirty number plates and shading on the characters of the number plate. The performance of the system can also be affected by detection rate of controllable factors like resolution of the camera, camera angle related to horizon, distance between the number plate and camera. Although by passing through all these problems, the authors gave the innovative criteria for the license plate recognition based on neural network, Fuzzy logic, Fourier transform, Genetic algorithms and wavelet theory.