ECE 153B Final Project

Facial Temperature Scanner

By: Geffen Cooper

Project Overview

This project uses the MAX78000FTHR development board to create a facial temperature scanner that reports if a person has a safe temperature. The scanner uses the on-board camera module and CNN accelerator to detect a person’s face in the image frame. The camera feed and instructions (‘move left’, ‘move up’, etc.) are displayed on an LCD to help the person center their face. Once centered, an IR temperature sensor will measure the person’s facial temperature which is then reported to on LCD

Block Diagram

This diagram shows the hardware layout of the system. See the parts page for more details.

Software Structure

This diagram shows the software control flow which is modeled by a state machine. See the project proposal or demo video for more details. The code can be found here: GitHub Repo. For details about the training and synthesis process of the CNN feel free to reach out! I am also very interested in pursuing this type of project (machine learning at the edge) for capstone so feel free to reach out if you are interested in that area as well.