Skip to content

Home

Jump to bottom
Travis Stewart edited this page Aug 30, 2018 · 1 revision

Computer Vision Emergency Response Toolkit

Project Purpose

This Project was developed as part of the Texas A&M University's 482 "Capstone Design" course. The purpose of our project was to develop an application that can aid rescue workers in locating survivors of natural disasters more rapidly. To that end, we developed a modular software package that currently implements two different algorithms, and applies them to a given set of images. The result of this software is a series of "heat-maps", which indicate possible features of interest within each of the given images.

Quick-Start Guide

GUI Application

  1. Run the Installer: Setup.msi
  2. This will install the program in AppData, and create shortcuts on the desktop for the Batches folder and a shortcut to the application. The Batches folder will be installed to your Documents folder for easy access.
  3. Run the application from the Desktop. This will begin the installation process.
  4. If Python 3.6.4 is not installed, the application will prompt the user to install it. Make sure that, when installing Python, "add to PATH" is checked.
  5. The application will install its dependencies, and then launch the program.

Back-End CLI

Running a new analysis:

python PythonScripts\analyze.py -F <Path_to_Image_directory> -p <number_of_threads>

Results will be stored in new folders inside of the directory you pass in using the -F parameter.

Software Design

The Design of the software was divided into two separate pieces:

  1. Python Backend
  2. C# frontend

The Front-end process simply executes the back-end using command-line arguments, which the back-end uses as the inputs, and as it produces results, the output is returned to the front-end using redirected standard output, which the front-end reads, parses, and uses to display the results.

Clone this wiki locally