Crowd CTRL: Social Distancing Drone Based System
Loading...
Authors
Rawan Al Musallam
Naqvi, Susan
Al-Sarabi, Mohammed
Hammoud, Hussein
Issue Date
2021-06-13
Type
Language
Keywords
ELEG/CPEG
Alternative Title
Abstract
Ever since the declaration of a pandemic due to the novel coronavirus (COVID-19), our world has been battling through difficult challenges. With too little known about the virus and its high mortality rates, lockdowns have been enforced by countries as a mean to limit and delay escalations in the number of infections. The sudden widespread of COVID-19 lies behind its high transmission rate. Consequently, it is crucial to be well protected during any form of contact. Early diagnosis and detection are vital to control and contain the virus. However, such task is not easily done especially when it comes to asymptomatic cases. Hence, pursuing approaches that aim towards the minimization of contact is one important factor of this project [1].
Procedures aiming towards early detection during the pandemic involved fever detection in patients. Medical professionals began seeking alternative ways to measure people’s temperatures. In the beginning, the medical staff used the old thermometers. Later thermometers were replaced by thermal cameras since their efficient nature requires no physical contact with the human body. Currently, different organizations are establishing various COVID-drone projects with aims such as delivery of medical supplies, spraying disinfectants and public surveillance [2], [3]. In spite of fever being the predominant symptom of COVID-19, even with the progression of thermal cameras, mass fever detection is not realistic as it is not possible to accurately detect fever in a crowd. The deployment of such inaccurate technologies can lead to a high number of negative results, hence exposing people to more unwanted risk [4]. Studies showed that in order to accurately detect fever, the forehead and inner corners of the eyes should be targeted for providing precise results. In addition, thermal cameras are pricey and are best used for individual testing.
While these technologies contributed to detecting fevers, one of the safest and effective means of controlling the spread of COVID-19 still remains as social distancing [4], [5]. Accordingly, we introduce Crowd Ctrl, a project that focuses on maintaining social distancing protocols specifically in closed areas susceptible to overcrowding. We aim to create an IoT based system able to manage the crowd density within a building by counting the number of people per area. The IoT device consists of a drone and a speaker controlled by a microcontroller for sounding an alarm. Concepts of Deep Learning and object detection algorithms are utilized for our system. Furthermore, the project will be cloud based, meaning processing will take place on the cloud instead of locally. The drone is controlled manually by the respective expert. Using the drone’s camera, a video stream is sent to the cloud where it is processed, if the amount of people exceeds the specified threshold, the microcontroller receives a command from the cloud to trigger the buzzers, hence alarming the crowd. Additionally, an application is built for users, allowing them to monitor the crowd level in a mall, hence, contributing to avoidance of overcrowding. Utilizing the knowledge gained throughout these years, we as engineers aim to find feasible, efficient, and improved measures to tackle down challenges encountered by others in this capstone project
Description
Citation
Publisher
College of Engineering and Applied Sciences