Driver\'s cell phone usage monitoring in a moving vehicle

Project Summary: In this project we will train our system to detect the driver cell phone usage in a moving vehicle. The input data of this project will be live video of camera installed on the road side. The camera will send the data to the system and t

Project Title

Driver\'s cell phone usage monitoring in a moving vehicle

Project Area of Specialization

Artificial Intelligence

Project Summary

Project Summary:

In this project we will train our system to detect the driver cell phone usage in a moving vehicle.

The input data of this project will be live video of camera installed on the road side. The camera will send the data to the system and then the system will detect the vehicle in video and then it will recognize the number plate of the vehicle and then recognize and identify the driver seat.

Now if the driver is using cell phone while talking or typing it will be detected and at the same time the number plate of the vehicle is also be detected and recognized.

Now the number plate of the car will be extracted from the image and compared with the data base. The data base will contain the information like car registration number, owner name, cell phone number etc. The driver will be challaned for traffic violation and notification will be send through mobile number.  

Project Objectives

Project objectives:

The main objectives of our project are:

• To make driving safer and prevent dangerous road accidents that causes due to the use of mobile phone while driving.
• The incentive for performing this project is to help improve driver safety on the road.
• Develop a system that automatically detects a person using mobile phone while driving.
• There are mostly indirect, non-visual based technologies to identify driver’s mobile usage which requires significant additional infrastructure and may need trained expertise to use. This project aim is to make a vision-based detection system which works directly on the existing traffic monitoring platform, with minor setup adjustments.

Project Implementation Method

Project Implementation:

• The mobile phone detection camera system incorporates a number of cameras and an infra-red flash to capture clear images of passing vehicles in all traffic and weather conditions.
• The artificial intelligence software automatically reviews images and detects potential offending drivers, and excludes images of non-offending drivers from further action.
• Images that are automatically deemed likely to contain a mobile phone offence will be verified by appropriately-trained personnel. Images rejected by the artificial intelligence will typically be permanently deleted within an hour of detection.

Benefits of the Project

Benefits of the projects:

Following are the main benefits of this project:

The incentive for performing this project is to help improve driver safety on the road. We view the costs and limitations of the current road user surveillance approaches as a factor in this issue. Hence, we attempt to use image processing algorithms to solve this.

Image processing algorithms have been used in other fields, such as in security and factory automation where thousands of parts and products are inspected with high speed, precision, and accuracy.

This image processing algorithm approach is not advocated as a total replacement, but to supplement existing CCTV systems and their users. Traffic law enforcers could use this approach to analyze large volumes of live CCTV feeds and even video recordings of drivers who are distracted by their mobile devices.

Technical Details of Final Deliverable

Technical Details:

• The artificial intelligence software automatically reviews images and detects potential offending drivers, and excludes images of non-offending drivers from further action. If no offence is detected, images are permanently and irretrievably deleted, typically within an hour.
• Importantly, the software is a screening tool only. If a possible offence is detected, there are then several stages of human review and adjudication before a penalty notice will be issued.
  • Images containing possible offences are verified by the vendor delivering the program. This check is completed by approved trained staff using a secure network. The images that are viewed at this stage are cropped and pixelated to remove information that would identify the vehicle or the vehicle location. Images that are do not contain offences at this stage are deleted within 72 hours.
  • If a likely offence is found in the first review of images by the vendor, files are securely transmitted to Transport for further review by trained officers. Images that are verified at this stage to likely contain an offence are supplied to Revenue .
  • Revenue conducts final adjudication and issues a penalty notices. If Revenue determines that an offence cannot be proven, then a penalty notice will not be issued.
• This process is similar to other camera enforcement programs in but with added human reviews to verify the potential offence identified by the camera system.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Transportation

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

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