Published Paper Details:

Increasing population has also increased the traffic graph.India was ranked number 1 among 199 countries for the most number of road accidents by world road statistics in 2018.So, a system is required to maintain and detect traffic rule violations. An automatic traffic rule violation detection and number plate recognition is an effective method and a step ahead in controlling traffic.ALPR is used to trap traffic violations and detect the vehicle number plate so that it becomes easy to identify vehicle owners. Different methods are used which include wavelet decomposition,image processing and template algorithm,spatial/frequency domain filtering and neural networks and DELP-DAR system. In this paper,we have used CNN - Convolutional neural networks.This is an efficient method to overcome the issue. The proposed system first captures the image of the license plate of the vehicle and the image undergoes several refining processes to produce a clear image of the number plate; these extracted characters are verified using CNN classifier. CNN is preferred because of its higher accuracy which is related to 99.43% in terms of localisation and 98.6% in terms of recognition.