In today’s times wearing a face mask has become compulsory to protect ourselves from COVID-19. Eventhough wearing a face mask has been included in the safety guidelines by the government yet most of the people don't follow it. This has led to tremendous increase in the covid cases. And since it is difficult to keep a track of the people not wearing masks hence we decided to develop this project.
What it does
The face mask detection system integrated with a CCTV surveillance camera identifies a human face and detects whether the person is wearing the mask properly, wearing the mask improperly or not wearing a mask and displays the message on the screen and sends a notification to the administrator if he is improperly wearing the mask or not wearing the mask.
How we built it
The technologies used by us are: 1)Platform: Jupyter notebook 2)Image processing and management tech stack: a) Open CV b) Numpy c) Matplotlib d) Scikit image 3)API: Sinch 4)Programming language: Python We used Haar cascade image classifier for human face detection, Support vector machine learning algorithm for classifying the human face into 3 categories: a)Properv mask b)Improper mask c)No mask Then we used sinch application programming interface for the purpose of sending alerts.
Challenges we ran into
1)Time constraint 2)Debugging the entire code. 3)Hinderance in discussion regarding the project with other team mates due to poor network connection.
Accomplishments that we're proud of
Our project is working successfully with an accuracy of 98.6 percent. This system can be utilised to keep a watch on people entering a particular place like banks, schools, malls etc without employing human work force and exposing them to infection.
What we learned
This project helped us in improving our technical and problem solving skills. This hackathon encouraged us to find a suitable solution for a real world problem. We also learnt how to integrate the views of each team member to bring out the best possible outcome.
What's next for Covid-19 Face mask detection system.
This project can be improvised by implementing voice signal processing. It would be able to detect if the individual is wearing a mask or not by analysing his voice.For example: If an individual is wearing a face mask then the intensity of his voice will be reduced(muffled) and this will be detected by the system.