Real Time Crowd Surveillance using Machine Learning and IoT

COVID-19 is an acute, sometimes severe, respiratory illness caused by the novel coronavirus. In 2019 First discovered in Wuhan, China, the virus has spread all over the globe, where places like Europe, India are facing the second wave of COVID19 pandemic in 2021. Now that the pandemic is nowhere to be seen to take an end, efforts must be made to mitigate the spread of the Virus. Organizations like Shopping malls, schools, Banks have turned out to be places where crowd gatherings happen the most, and here the chance of being infected also increases. So, to stop the spread of the virus, we have proposed a system that provides smart surveillance in crowded areas. In the proposed model we have done Face Mask detection, temperature Monitoring along with continuous tracking of crowd density and to integrate all these modules we have used Multiple Regression Mode. For Mask detection, we have used the OpenCV library called Dlib, which accurately extracts the face feature and with specified conditions, it can be identified if a person is wearing a mask or not. For monitoring of temperature, MLX90641 is used, with a pi-cam to get the video footage, along with keeping track of people entering. All the Required Computation is done on the Raspberry Pi and according to the specified condition, the Raspberry pi signals the servo motor to close and open the entry door.