A Novel Cough Detection Algorithm for COVID-19 Surveillance at Public Places

Sreedharan, D.; Subodh Raj, M. S.; George, S. N.; Ashok, S.

A worldwide pandemic, COVID-19 has been caused by a newly discovered strain of coronavirus SARS-Cov-2. Its common symptoms are high fever, coughing, and shortness of breath. With the rising number of COVID-19 cases, manual detection of infectious individuals at public spaces is a hectic task. Artificial Intelligence (AI) based detection systems can be deployed at public places like airports, railway stations, etc. for continuous monitoring of potential infectious individuals and screening based on common symptoms exhibited. In this paper, a new algorithm is developed for detecting repetitive coughing action which is the main symptom in COVID-19 cases, and thus detecting people with COVID-19 based on it. The performance of the proposed system is tested on an existing sneeze-cough dataset and also on a real-Time dataset. The evaluation shows that the proposed method has superior performance over the state-of-The-Art methods. © 2021 IEEE.