• Gaurav Jaiswal
• Aditya Tiwari
• Siddharth Bhadauria
• Justine Kizhakkinedath
• Anil Bhadgale

Real time human detection and tracking on a drone under a dynamic environment is the key technique in the field of intelligent transport. A new lightweight real-time onboard human tracking approach with multi-inertial sensing data is proposed. The first thing built is the drone and its hardware consists of an embedded system used as the flight controller Ardupilot. The peripherals include brushless direct current motors, electronic speed controller, frame and radio receiver-transmitter, and ultrasonic sensors and GPS module. The flight controller calculates the yaw, pitch and roll of the drone at equal intervals and by using PID values it calculates the corrections to be made to the drone to keep it stable. Another more powerful embedded system Raspberry Pi is also used to carry out the smart features of the drone such as image processing and taking decisions to track the human in frame. Drone would be able to do feature of human detection and tracking using image processing. ssd_mobilenet model is used for image processing. The user interface consists of a Radio Transmitter and a web interface, for live streaming of video being captured by drone and target selection.

Unmanned Aerial Vehicle (UAV), Machine Learning, Object tracking, Object detection, Computer vision.

"Smart Drone Implementing Detection And Tracking Of Humans", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.720-723, April-2019, Available :http://www.jetir.org/papers/JETIR1904198.pdf

"Smart Drone Implementing Detection And Tracking Of Humans", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp720-723, April-2019, Available at : http://www.jetir.org/papers/JETIR1904198.pdf