2020 has been a testing year for all of us. It has changed our lives drastically. Going outside your home now means securing your mask, frequently reminding yourself of the term “social distancing” and getting your temperature checked as soon as you enter any public premise.we have devised a solution for all industries and crowded areas where security and safety issues are a concern. By developing an economical and automatic system that is capable of measuring the body temperature of a person combined with Face-Mask Detection and Facial Recognition, a company will be efficient in carrying out an additional screening before the employee with initial/potential symptoms could enter the location.
What it does
M-GAP does the following things:- i) Measure body temperature (initial COVID symptom). ii) Detect mask equipment. iii) Perform facial recognition. iv) Generate an alert according to the favorability of the outcome. v) Monitors and records each detail about an individual for future use minimizing manual work.
How we built it
The project utilizes OpenCV and Tensor Flow open source libraries to perform Facial Recognition and Face-Mask Detection. Using Flask, a web framework built atop Python and WSGI, we have built our back end. The whole solution performs the functions listed above by taking an input from the web camera and the rest through external inputs. The team designed its database using PostgreSQL due to its powerful architecture and performance.
Challenges we ran into
The project in itself is a challenging task to complete as we aren’t previously familiar with the technology we will be using. We are new to implementing facial recognition and face-mask detection separately and now since we are aiming to build it in a combined state, it makes it more challenging. The accuracy by testing them is a demanding task. We are unsuccessful in creating a hardware device due to unavoidable reasons, which makes us get onto a decision of creating a web application performing the exact same features. Linking all the separate applications into a single web application and making it work and get a favorable result requires effort and is challenging enough.
Accomplishments that we're proud of
We are proud to have completed the project in a limited time frame with full efficiency by testing it multiple times. Additionally, we have tried to keep a user-friendly interface design on such a complex back end which in itself was a huge accomplishment.
What we learned
The journey of building the project was a great learning experience for the team as we had first time collaborated to build a project together. Furthermore, the team gave in exceptional efforts to learn and implement machine learning algorithms into the project idea.
What's next for M-GAP
Implementing our web application into a hardware device is a future update we would be working on. Thus, with the use of thermal sensors, we will be creating a fully non-contact device which will be a huge innovation in our project.