This is a capstone project for CSEC-490 Capstone in Computing Security, for CSEC students at Rochester Institute of Technology. The goal of this project was to implement biometrics in the form of retinal authentication for a mobile device, which can be readily accessible for the general consumer to use. It does so by engaging the camera on certain speciality phones, particularly ones with a built-in IR camera, to detect the user's eyes and retinas. The user is then prompted to visually draw a pattern on the phone's screen using their eyes, similar to the swiping pattern on Android devices. If the pattern is correct, the user is authenticated into the device if they are a valid authorized user.
Pre-requisites:
- Python 3.7+
- Android Studio
- Mobile device with IR camera (ie. Galaxy S7)
Ensure that the opencv-master folder is in the same directory as that of the Python script
Push the application from Android Studio onto a supported phone with a built-in IR camera.
OR
Download the Python script onto any directory on your computer and run the script from IDLE or any Python IDE.
- Current iteration does not accurately detect retinas
Button_Authentication.py
- Purpose: Provide a method of retinal eye authentication by allowing the user to input a password sequence through covering their eye over points on the GUI for 5 seconds to confirm their input than compare their password against the current password
- How to use: Run the program with Python and hover your mouse cursor over each button for 5 seconds, once you have hovered over the button for 5 seconds move it off and the program will accept it as input. Once a sequence has been inputted hit the 'Authenticate' button to test against the current password
- State: Currently in a fully functioning testing state
- Integrate all applications into one singular application
- Fine-tune retinal detection capabilities
- Implement a database to store previously authenticated users
- Create a GUI for all available capabilities on the application
Utilizes OpenCV as an open source library for a majority of the application functions and Android Studio as the primary IDE.
Special thanks to our sponsor IPPSec and our faculty advisor Bo Yuan.
Jhony Alavez, Computing Security, jia2707@rit.edu
Nicholas Laufer, Computing Security, nxl6264@rit.edu
Nicholas Kurland, Computing Security, ntk1002@rit.edu
Benny Tan, Computing Security, bt4168@rit.edu