Machine Learning based Mechanism for Crowd Mobilization and Control

Abstract

At present, the world is in a difficult pandemic situation i.e., COVID-19. More than 95 Million cases are found and are increasing every day. Greater than 2 Million people died of the virus. Currently, The United States, India, and Brazil are the top most affected countries, and many countries are working to discover the vaccine. After successive lockdowns, the world is trying to come back to normal life with safety restrictions. So, every place like malls, schools, shops, temples, etc., is mandatory to check the temperature and mask worn by every person. The process is time-consuming and involves large human intervention thus increasing the expenses. Hence a setup is to be developed that detects face masks, measures temperature, and controls the crowd entering the building. To achieve these operations, an IR-proximity sensor, IR-temperature sensor, a barricade system, a microprocessor, and LED indicators are used. The setup is fabricated using low weight, low-cost PVC pipes. At first, the proximity sensor checks whether the person is entering the setup and gives input to the microprocessor Raspberry pi which then opens the 1st barricade. Temperature sensor works on the principle of Stephen-Boltzmann law. The temperature of the person is measured and the pi camera detects the face mask. The face mask detection is done by the microprocessor using Open CV and Tensor Flow by comparing images provided in a dataset enhancing machine learning methodology. The entire process is powered by a 12V DC adapter. The barricade system can be installed at all major public places and the system functions effectively.