Overview / Usage

Novel AI Machine Learning Intel OpenVino Edge Computer Vision System consisting of Thermal Imaging Camera with Infrared non-contact Temperature Sensor that detects humans, faces and their body temperature with Machine Learning Edge Computer Vision in real-time authorizing virus free. Using Intel OpenVino and Pre-Trained models creating python and C++ apps which will run real-time Machine Learning Inference on Facial Detection, once face is recognized with Computer Vision and Webcam or Flir Camera, then Flir Camera is activated which is a thermal camera that also detects pinpointed Temperature and a range of thermal imaging temperatures, code will include that any temperature that is too high and recognizes a fever range will alert and notify operator. Device will be used in all businesses and retail in USA. May add on Intel RealSense for real-time Machine Learning Distance Inference data. All data will also be sent to the cloud. May use Raspberry Pi 4G system or Laptop system. Using Intel DevCloud to Train Machine Learning Models.

Methodology / Approach

Frameworks C++ and Python with Intel OpenVino and Flir Camera with Flir SDK. Pre-Trained Machine Learning models for Intel OpenVino real-time inference. Code for facial detection, if face detected will activate Flir and other Temp Gun, other Temp Gun may be robotically controlled. Flir will read temperature and thermal imaging range, will notify operator via app and or cloud MQTT and Node.JS Web Server if Temperature is too high, if so, will be alerted and denied entry of business or retail and escalated to further symptoms. Will also do facial authentication so that if person returns to enter and has already been in range will be denied entry if still in high temperature range.
I have the Facial Detection done, I have Flir Thermal Imaging Camera coming, I won the Intel Edge AI ML Udacity Scholarship and now in new Nanodegree program and will be doing this project as one of my projects in the Intel AI ML Edge Nanodegree program starting end of April 2020. Using Intel DevCloud to Train Machine Learning Models.

Technologies Used

Intel OpenVino, Pre-trained Machine Learning Models, Intel RealSense Camera, Raspberry Pi, MacBook Pro, Windows Laptop, Linux laptop, Intel Neural Computer Stick 2, Web Camera, Flir Thermal Imaging Camera. Flir Temperature Lenses, Python, C++, CMAKE, Node.JS, MQTT, FFMPEG Server. Using Intel DevCloud to Train Machine Learning Models.