UDriveSafely

Posted on by jxg4185

Team Name

The Achievers

Product Name

UDriveSafely

Timeline

Summer 2019 – Fall 2019

Students

  • Bimal Pathak
  • Roshan KC
  • Yubaraj Bhatta
  • Bimal Shrestha
  • Saroj Lamichhane

Abstract

With the increasing number of traffic, and also people being more addicted towards use of cellphone or
other technological devices while driving, has made our roads less safe. People, being unaware of the
surroundings and traffic signals has led to the increased number of road accidents. Majority of people are often get confused with the similar looking traffic signals and without knowing what to do in such kind of traffic signals, it often leads to the road rage  and possibly an accident. Therefore, we come
up with a project that will help people  be aware of the surroundings while driving on the road. Our
project is solely an android application which detects and make people conscious about the signal
and lights. It will decrease the rate of an accident happening now and improve the quality of driving
skills with the notification alert it provides to the user when detecting the traffic signals ahead of the driver .

Background

As reported in the findings of an eight-year study conducted by the National Highway Traffic Safety Administration (NHTSA), there were on average 1,578 fatalities each year resulting from two-vehicle traffic crashes at intersections controlled by traffic signals. Approximately 51% of those fatal crashes were caused by drivers who ran red lights. Approximately 29% were caused by drivers who failed to yield the right-of-way at traffic signals. Because of such kind of issue some developer comes with different software which really can make people aware of the situation. But, some of the software has high-cost people are not able to purchase and use it. On the other hand, some of them are not longlasting and the system has not been working normally like before. Since we are trying to develop this app with low cost and sustain for a long time, it will be the best option for the customers. Therefore, the app will help the user to prevent unwanted accident.Our goal is to create a userfriendly GUI that interacts with users and help them through the app. Some of our features will have detection and alerts. For our product, we plan on making it affordable to those with lower incomes. We hope to save people’s life from this product. This product will be competitive with similar products on
the market with quality and affordable price.

Project Requirements

  1. The system shall run on Android phone only
  2. The application shall work on Android version with API level 23 or higher
  3. The application shall start with in 2 seconds.
  4. The System shall capture the video and detect traffic signals
  5. The application shall use the default camera built within the cellphone system
  6. The application shall detect the traffic signals very fast under good circumstances
  7. The application shall warn the users in timely manner.
  8. The Doxygen and GitHub shall be used for source code documentation and version control.
  9. There shall be no maintenance and support to the application in the future
  10. Future teams shall be provided application website for technical documentation

System Overview

Our system overview comprises of four layers: Input layer, Application layer, TFLiteTrafficSignalDetectionAPI layer that works as a Control layer,and a Presentation Layer. The primary tasks of Input layer is to get the frames from the video frame of the camera, convert those frames to appropriate image format that can be process by our object detection model. The application layer is responsible for getting the input frame from the input layer and then process the image. Processing the image is done by sending the image to the TFLiteTrafficSignalDetectionAPI to recognize if any signals are present in the frame. The TFLiteTrafficSignalDetecting API layer after processing the image sends back the recognition back to the application layer. Then, Application layer processes the recognition like ignoring detection below the minimum confidence level and then finally, alerting the user using the Presentation Layer.

Demo Video

 

Future Work

 

Project Files

Project Charter (link)project_charter_latex

System Requirements Specification (link)system_requirements_specification_latex

Architectural Design Specification (link)architectural_design_specification_latex

Detailed Design Specification (link)detailed_design_specification_latex

Poster (link)poster

References

 

jxg4185

More Posts