Campus Safety and Hygiene Detection System using Computer Vision

The recent spread of severe acute respiratory syndrome coronavirus 2 and its associated coronavirus disease has caused extensive public health concerns. University campuses are at higher risks since a lot of students are present inside the campus at a given point of time. Places where there are a lot of chances of spread of the infection in the campus include the entrance gate, canteen, library, photocopy center, seminar hall, etc. Strict actions must be taken against the violations of the covid-19 protocols which will ensure health safety and maintain hygiene in the campus. Doing this manually will be a tedious task. Owing to this problem, an attempt has been made to design a system to tackle the problem of following all the protocols and making everyone aware about the situation in the campus. This work proposes a system which will continuously monitor all these activities with the help of Computer Vision and Deep Learning. The collected CCTV cameras data has been checked in the real time mode using various object detection and object tracking models to identify and track the objects visible in the frame. This approach uses MobileNet and SSD Architecture along with the objection detection models to predict the desired output. Finally, based on the output the system checks for any violations and if encountered then it sends a text alert to the concerned authority.