Automated Traffic Signal Penalty System using IoT and Machine Learning

  • Gauri Shenoy et al.


In the current developing nation with an increase in population there is an increase in the number
of traffic rule violations. Maintaining these traffic signal rule violations has always been a monotonous
and time-consuming task. Although the current traffic management system is automated, the increasing
population and use of vehicles and the diversity of number plates makes this task more difficult. The
primary objective of this paper is to control these traffic rule violations precisely and cost friendly. The
proposed paper includes an automated system which uses ultrasonic sensors and a camera which is
connected to Arduino to capture video and take snapshots of vehicles. This paper presents automatic
recognition of number plates of vehicles that mainly cross the pedestrian crossing when the signal is
red. This could be done by implementing various machine learning techniques and image processing
techniques for number plate detection and character recognition making this task much quicker and
easier to identify number plates. Once the vehicle number is recognized from the number plate, a SMS
will be generated on the registered mobile number on the number plate stating the details of the traffic
rule violation that has occurred. This paper proposes a more cost effective and efficient automatic
system to ensure more safety on roads especially near the pedestrian crossing.