Electronic Toll Collector Framework

The project aims to develop a system that reduces the waiting time at the toll plazas. The existing system is very old and needs to be enhanced according to cutting edge technologies. We propose an Artificial Intelligence-based toll collection system which can issue toll, keep track of the toll in a

Project Title

Electronic Toll Collector Framework

Project Area of Specialization

Artificial Intelligence

Project Summary

The project aims to develop a system that reduces the waiting time at the toll plazas. The existing system is very old and needs to be enhanced according to cutting edge technologies. We propose an Artificial Intelligence-based toll collection system which can issue toll, keep track of the toll in a database, get payments from drivers via phone/postpaid-payments/payment to cashier (for drivers without phones/dead-phones). It also stores the timestamp and vehicle type plus an image of the vehicle that has passed. The project can help the law enforcement agencies to check for any particular vehicle and to keep track of the toll collected by the plaza. The system uses Google’s tensor flow for vehicle classification. The software can perform classification on images, saved videos and live streams. The primary feature is to perform all the above features and save the information in the database, in real-time.

Project Objectives

The primary objective of this project is to develop a system capable of doing most of the tasks that human do manually and often inaccurately. The software system should be able to

• Real-time video-based vehicle classification and data insertion in the database.
• To generate the toll receipt for all type of vehicles.
• To maintain the record of toll collected, on daily, weekly, monthly and yearly basis.

To create a website for toll analytics, accessible by the government officials.

Project Implementation Method

Step 1: Data Gathering

First task would be to build vehicle dataset solely for the vehicles that are used in Pakistan. All the images would be in .jpg format.

Step 2: Vehicle Classification Using Deep Learning

Convolutional neural network (CNN) will be trained and tested on our dataset.

Step 3: Generating Toll

Based on the classified type of the vehicle, system will generate the toll. The driver would be able to pay the toll via phone/monthly billing or by hand to the cashier in case he/she don’t want to use the service.

Step 4: Development of Web Portal/Interface

The information will be stored in the local system running the software, which will be then sent to a centralized server to get daily/monthly/yearly analytics on the web portal and the analytics would be available to higher government officials.

Benefits of the Project

The local or federal governments and law enforcement agencies can be the ultimate beneficiaries. Following are the benefit offered by the project: 

1. Reducing the man power
2. less expenditure
3. Automated toll collection - no human negligence or error
4. Daily/Monthly/Yearly Analytics on the web.
5. Transparent Finance

Technical Details of Final Deliverable

The final deliverable will be a high definition Camera and a computer system with a GPU. The system will need Wi-Fi enabled so that it can send the information back to the centralized server. The server would save the information in database and perform analytics on the data. We will need a MySQL open-source database and website hosting. Images will be stored in the local file system of the computer. The camera would be connected to the system for live videofeed. We will also have a toll printer as a deliverable to print the respective toll.

Final Deliverable of the Project

HW/SW integrated system

Type of Industry

Finance , Transportation , Security

Technologies

Artificial Intelligence(AI), Internet of Things (IoT)

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com