A hybrid deep learning based approach for the prediction of social distancing among individuals in public places during Covid19 pandemic

ABSTRACT

Social distance is considered one of the most effective prevention techniques to prevent the spread of Covid19 disease. To date, there is no proper system available to monitor whether social distancing protocol is being followed by individuals or not in public places. This research has proposed a hybrid deep learning-based model for predicting whether individuals maintain social distancing in public places through video object detection. This research has implemented a customized deep learning model using Detectron2 and IOU for monitoring the process. The base model adapted is RCNN and the optimization algorithm used is Stochastic Gradient Descent algorithm. The model has been tested on real time images of people gathered in textile shops to demonstrate the real time application of the developed model. The performance evaluation of the proposed model reveals that the precision is 97.9% and the mAP value is 84.46, which makes it clear that the model developed is good in monitoring the adherence of social distancing by individuals. [ FROM AUTHOR] Copyright of Journal of Intelligent & Fuzzy Systems is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)