On Stream Face Mask and helmet detection System Using Machine Learning Techniques

Within the context of the current day, safety and cleanliness are the most important factors to consider in order to guarantee a healthy and risk-free human existence. Research has constantly proven that basic hygiene is vital, and that wearing a mask in public areas is crucial, in light of the terrible epidemic caused by the Coronavirus. Consider mandating the use of helmets as another precautionary safety measure for users of two-wheeled vehicles. A real-time Yolov3 object detector is used to train the model to recognise face masks and helmets in video footages, live feeds, or photos in order to search for public hygiene and safety. This is done in order to ensure that people are not exposed to potentially hazardous conditions. A neural network is used for the prediction of bounding boxes and the possibilities associated with those boxes over the whole picture. The dataset was gathered via the use of the internet. The model has been taught with the use of internet photos for several classes, such as wearing a mask but not a helmet; not wearing a mask or helmet; wearing a mask but not a helmet; wearing a helmet but not a mask; and wearing both a mask and a helmet. The trained dataset is used mostly for the purposes of classifying items and locating them within the aforementioned categories. In the event that neither the helmet nor the mask is located, the SMTP library module that is supplied by Python specifies an SMTP that may deliver mail to any internet device using SMTP or ESMTP as a client session object. The suggested approach was evaluated using a dataset that included photographs of persons adhering to COVID-19 safety regulations as well as photographs of people adhering to road safety recommendations while wearing helmets. Accuracy and precision are both quite good in the outcomes that were achieved.