Intelligent Traffic Monitoring System
We propose a deep learning model to control traffic lights linked through IoT. The proposed model is composed of several components to improve the performance such as congestion control, number plate scanning and automatic street lights using sensors and raspberry pi to integrate whole setup. Our de
2025-06-28 16:33:20 - Adil Khan
Intelligent Traffic Monitoring System
Project Area of Specialization Artificial IntelligenceProject SummaryWe propose a deep learning model to control traffic lights linked through IoT. The proposed model is composed of several components to improve the performance such as congestion control, number plate scanning and automatic street lights using sensors and raspberry pi to integrate whole setup. Our developed solution is targeted to incorporate the progress in the internet of things (IoT) and deep learning, where low power, embedded devices integrate as part of next-generation TMS.We propose a deep learning model to control traffic lights linked through IoT. The proposed model is composed of several components to improve the performance such as congestion control, number plate scanning and automatic street lights using sensors and raspberry pi to integrate the whole setup. Our developed solution is targeted to incorporate the progress in the internet of things (IoT) and deep learning, where low power, embedded devices integrate as part of next-generation TMS.
Project Objectives1)Counting of cars using deep learning models and performing several functions upon counted cars according to the situation of congestion.
2)Scanning of number plates of the vehicle in order to check upon violations.
3)Automatic street lights for power saving purposes.
Project Implementation Method1)First training of model and counted cars through object detection using CNN has been performed
2)Performed checks on stored counting vehicle variable for operating of signals
3)Integration of raspberry pi and trained model using the camera
4)Training of number plate detection for challan violation
5)Integration of raspberry pi and trained model of challan violation
6)Integration of both models for 2-way intersections
7)Using the wood model and embedding automatic street lights using sensors.
Benefits of the Project1)automatic control on congestion of traffic.
2)Smart signals are developed having human behavioral qualities using deep learning
3)The cost of the whole TMS is decreased by 70%.
4)Low processing power is required
5)Due to automatic street lightning, power saving is 47% than normal lights
6)The cost of labor is saved as no labor or wardens are required on signals.
Technical Details of Final DeliverableThis system is designed to govern traffic at road networks, sensing through cameras by
means of Deep Learning , surveillance IR sensor which is embedded on roadsides in
order to save energy on highway lights. The system works in a distributed manner, it
processes the cameras that operate the Signals control and the Detection of Violation
of traffic rules monitored by a raspberry pi.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| CPU+GPU | Equipment | 1 | 50000 | 50000 |
| RASPBERRY PI | Equipment | 2 | 7500 | 15000 |
| ARDUINO MEGA | Equipment | 1 | 2500 | 2500 |
| IR SESNORS | Equipment | 10 | 250 | 2500 |
| MODEL | Miscellaneous | 1 | 10000 | 10000 |