About the project:

This project is a security surveillance program designed to used deep learing facial recognition to create a database of captured faces to be searched through by law enforcement to identify and track down criminals. The idea is for it to be fed via security cameras. The program is designed for 16.04 LTS

Here is a demo of the application: FF84™ Webpage

How to use FF84™

FF84 is a simple to understand program user side; to begin the search through the database a user simply must use either of the New Case links (Pictured Below)

New case image

Then the user must input the image of the criminal (pictured below)

Case creation screen

against the database, the program will then output whether or not there was a match as well as the times and locations that the suspect was detected.

Project Functions

  1. Stream.py > This function reads the frame by frame input from our camera(s) and sends the captured images to the server.py
  2. Server.py > Takes in images from Stream.py and then passes them to Facecut.py
  3. Facecut.py > This function takes the image inputs from Server.py and uses facial recognition to cut out the detected faces and > after writes them to the mySQL database for later comparison
  4. Facematch.py > This function does the comparison of the face entered on case creation across all faces in the database, then > outputs whether or not there was a match and the timestamps and locations for each positive match

Project Requirements

PHP Python3 pip3 pillow face_recognition

Project Developers

  • Alex Levanthal
  • Bennett Warner
  • Ellis Madagan
  • Gavin Macko

remember, if you have a face, we will find you.

Built With

Share this project: