Crossing Buddy Project Plan

Members

Alex Esenwein - aesenwein2018@my.fit.edu

Ryan Schwieterman- rschwieterma2018@my.fit.edu

Joshua Quinto - jquinto2018@my.fit.edu

Elias Anastasopoulous - eanastasopou2018@my.fit.edu

Faculty Advisor: Dr. Mitra - dmitra@fit.edu

Client and meeting dates

Client: Dr. White - rwhite@fit.edu

​

Meeting: September 7th, 2021

Goal and motivation

• The overall goal of this project is to create a system to help the visually impaired cross the street in a safer way.
• Audio cues at crosswalks aren’t standard so the blind have to listen for cars which has a small margin of error.
• With the rise of electric cars this listening method will become less effective and more dangerous.

​

Key features

• The system will be easily portable for the user's convenience.
• This will be achieved using light weight batteries and a vest that can be removed easily.

Key features

• The next key feature will be the ability to see and understand the crossing signal.
• The camera will identify the crossing signal and notify the user when they have the right of way.
• The system will communicate with the user via audio cues ex “you have the right of way”.

Key features

• The system will be able to detect cars on the users left and right.
• Using cameras mounted on the shoulder the system will notify the user of oncoming traffic.
• The system will communicate with user via audio cues ex “incoming traffic”

Novel features

• Uses You Only Look Once to detect cars and crosswalks simultaneously
• The different instantiations of YOLO will work in tandem to give the user a 270 degree view of incoming cars and crosswalk lights

Technical challenges

• The first technical challenge is using computer vision to identify objects. No one in our group has much experience with computer vision.
• Another technical challenge is the hardware components for the system. Our group has mostly software experience so we will have to learn about hardware through the course of this project.
• The final challenge will be creating a computer system on the vest that will allow the code to use the cameras to analyze. We plan on using a vest that will hold cameras on the shoulders and chest to get a 270 field of view but the problem lies with having the whole system connected.

Milestone 1

• Compare and select cameras to use in the system
• Select how the cameras will connect to processor
• Select where we will get the vest for the system
• Create Requirement Document
• Create Design Document
• Create Test Plan
• Demo technical tools

​

Milestone 2

• Implement the streaming to computer
• Have hardware components connected
• Begin testing image processing with test images
• Implement, test, and demo feature

Milestone 3

• Begin testing crosswalk detection feature in real situations
• Implement, test, and demo feature