Publication Type : Conference Paper
Publisher : IEEE
Source : 2019 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET) Pages 335-338, 2019
Url : https://ieeexplore.ieee.org/document/9032876
Campus : Amritapuri
School : School of Engineering
Department : Electronics and Communication
Year : 2019
Abstract : Since the past decade the use of autonomous and semi-autonomous robotic systems for human assistance has gained immense significance in the fields of defense, medical surgery, industrial automation, automobile assembly, etc. In particular, the development of Unmanned Aerial Vehicles (UAVs) has been a boon to human kind because of their agility to survey inaccessible places with ease. Consequently, UAVs could provide prominent and multi-faceted information in a Search and Rescue (SAR) operation. In order to mitigate the number of casualties involved in a natural disaster, rapid localization of the victims involved is of utmost priority. The paper in hand elucidates the development of an automated human detection system mounted on a drone that can provide the rescuers with the vital data required to execute an effective rescue mission. The proposed UAV can autonomously carryout survey over a preplanned region of interest and detect the survivors. It is also capable of assigning a score of confidence for each human detection, from the video captured through the camera mounted on the UAV. Consequently, this would help in drastic reduction of time spent in coming up with an action plan to accomplish the rescue operation. After a few successful trials, results have been acquired with moderate accuracy in detecting humans with a higher level of confidence.
Cite this Research Publication : S. L. Krishna, G. S. R. Chaitanya, A. S. H. Reddy, A. M. Naidu, S. S. Poorna and K. Anuraj, "Autonomous Human Detection System Mounted on a Drone," 2019 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET), Chennai, India, 2019, pp. 335-338, doi: 10.1109/WiSPNET45539.2019.9032876.