Traffic signal violation monitoring through video surveillance using CNN

Computer Engineering

Most of the traffic jams and accidents are occurred due to the bad traffic rule violation habits of the drivers or riders. This often leads to chaos and it may cause serious injuries, sometimes death also. This type of offenses can curb by identifying the offenders immediately and punish them with hefty fines. But many times due to over traffic and extreme weather conditions, traffic handling police personnel cannot do their duty up to their threshold. Availability of the artificial intelligence system which can do identify the offender and fine them immediately can bring the problem to considerable low level, this will lead to smooth traffic. Some methodologies are existed to identify the vehicles using Radio frequency identification tags (RF ID), this is again a costlier affair as all vehicles are not equipped with RF ID tags. So to better the process of traffic violation offender detection proposed model uses web cameras to identify the violated vehicle plate number through Convolution neural network, which uses the pattern matching technique for identification of numbers, thereby to fine the offender for the same and informing him through a text SMS.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.