Headcounter - Object Detection Software

A program for tracking people entering and exiting a building that additionally provides useful functionality such as setting a maximum store capacity (i.e. for enforcing COVID-19 safety measures) and sending notifications to involved parties.

Our project, Headcount, provides a solution for store owners to monitor customer activity in their store. By utilizing object detection software, Headcount tracks the number of people that are entering and exiting a building at any given time, notifying all involved parties in the case that the maximum capacity is reached. The program provides a user friendly interface and customizable experience allowing store owners to control settings such as maximum store capacity and providing a portal to communicate with customers and staff via email. The program also uses the data it collects from the object detection software to provide store owners with useful statistics regarding customer traffic and behavior within their stores.

Table of Contents

Requirements

  • CMake (version 3.1.0 or later)
    • On linux, you can run sudo apt install -y cmake.
  • Qt5
    • On linux, you can run sudo apt-get install qt5-default.
    • More information available here.
  • OpenCV4
    • Follow the OpenCV linux installation tutorial found here.
    • However, for this program, OpenCV needs to be compiled with the flag WITH_QT=ON.
    • You can use the cmake command found here for the cmake portion of the installation process.
    • Additional instructions are provided in the opencv_installation.txt file.
  • SimpleMail2Qt5

Compilation

To compile this program, run the following commands from the root of the project:

cmake .
make

Testing

To test the program, run it from the command line using the following command:

./headcount