MICROANALYTICAL’s Face Mask Detection is an AI and Computer Vision driven image analytics solution which caters to the Covid-19 related violations. It’s artificial intelligence program detects violations like Face Mask Detection, Social Distancing Monitoring etc.This system can be deployed on the Hospitals , Office Premises , Government Offices , Schools and Education Institutes, Construction sites, Manufacturing units, Airports etc.
The camera with AI-based face mask detection can generate real-time alerts.
Face mask detection feature uses visible stream from the camera combined with AI techniques to detect and generate an alert for people not wearing face masks. A user-friendly interface allows monitoring and review of alerts generated by the system.
Methodology
Neural Network created for Mask Detection System is based on Object Detection.
Object Detection
It is a technology that detects the semiotics objects residing inside a class in the form of digitized images and videos. One of its real -time application is self-driving cars. In this, our task is to detect multiple objects from an image. The most common object to detect in this application is the car, motorcycle and the pedestrian. For discovering objects in the image, we use Object Localization and it can identify multiple objects in real time systems.
The more complicated problem, of object detection involves both classification and localization. In this case, the input to the system will be a image, and the output will be a bounding box corresponding to every objects residing an image, along with the class of object in each box. An overview of all these problems is depicted in the below figure.
We have used YOLO (You Only Look Once) model for detection of person wearing face mask. YOLO does the training on full-fledged images and straightly optimizes detection performance.
Our Artificial Neural Network uses distinctive features from the whole image to predict and create each bounding box. It also predicts every bounding box in utmost different classes for an image sequentially. The YOLO depiction allows throughout training and real time speeds while nurturing higher average precision in every go through.
Linear Regression algorithm is used for training of our model to predict accurately. Model has been trained on thousand of images splited into test and validation folder. Test contains 80% of whole dataset and validation consist of 20% images from the dataset.
Visualization & Analysis
This analysis shows no. Of persons wearing or not wearing mask on the basis of “Date & Time”. 
The above pie chart shows percentage
Of people’s who are wearing mask on the basis of specific “Location”
With the help of general camera we have analyzed few videos .For example we can in first visualization on 27th April 2020, at 13 hours (1 PM) there are 28 people who are wearing mask and 11 are without mask .
Whereas , the second visualization depicts the pie chart where it shows the location wise information about people wearing mask i.e in old avenue there are 497 people who are wearing masks.
Key Features
Automatically Sends Alert
Send alert to people whose faces are detected without mask, also set the rate of sending the alerts and detection of faces.
Multi-Channel Recognition
Attach multiple cameras in a few minutes and enable all the cameras to access the AI capability of recognizing faces.
No new Hardware to Install
The system can work on any existing RTSP(Real Time Streaming Protocol) camera without the installation of any new cameras. Most of the hospitals and airports have IP cameras installed and RTSP-enabled.
How it works
Face Mask Detection Platform uses Artificial Network to recognize if a user is not wearing a mask. The app can be connected to any existing or new IP mask detection cameras to detect people without a mask. App users can also add faces and phone numbers to send them an alert in case they are not wearing a mask. If the camera captures an unmasked face, a notification can be sent out to the administrator.
If the face mask detector application identifies a user that he/she was not wearing a mask, AI alerts are sent with the picture of the person. It allows the application to run automatically and enforces the wearing of the mask.
