|Skill Level||Area of Focus||Operating System||Platform/Hardware|
|Intermediate||Computer Vision, Embedded, IoT, Security, Smart Home||Linux||DragonBoard 410c|
Autoknoby is a facial recognition system based on the DragonBoard™ 410c development platform from Arrow Electronics that is designed to authorize approved users to access physical control of certain household appliances. It is a modular smart device that can be installed on stovetops. Once an authorized user is detected by the system, it allows them to change the state of the physical control. This technology allows us to inexpensively convert an existing device with a physical user interface into a smart, mobile and Internet-of-Things device.
Seven months ago, in the Bronx area of New York, an unattended 3-year-old boy was playing with a kitchen stove which caused a fire resulting in 12 deaths and 6 critical injuries. Home fires such as this one cause an average of 2,500 casualties every year. Furthermore, kitchen fires make up 50 percent of all apartment fires and they’re the leading cause of fire injuries.
The desired outcome is to create tamper-proof physical controls that prevents loss of property and life. Furthermore, we hope to accelerate the adoption of Internet-of-Things devices with our easy to use technology.
Build / Assembly Instructions
- Attach the Relay to the Mezzanine Board by connecting the positive lead to Mezzanine ground, and the negative lead to the ground of the servo. Connect the relay signal to a GPIO pin on the Mezzanine board.
- Connect the RGB LCD to the LCD Mezzanine board.
- Connect the servo PWM signal pin to the Mezzanine board’s PWM pin.
- Connect the webcam to the DragonBoard 410c’s USB port.
- Connect the monitor and respective power supplies for the DragonBoard 410c and monitor.
- Connect the keyboard to the DragonBoard 410c by using the USB receiver.
- Boot into Linux and continue to the Usage Instructions
- Open Terminal
- Ensure Python is installed, if not, install Python
- Install face_recognition Python library ‘pip install face_recognition’
- Install cognitive_face Python library ‘pip install cognitive_face’
- Install OpenCV Python library ‘sudo apt-get install python-opencv’
- Download the code from github and execute the script using ‘python smart_dial.py’