Demo of web-app
While you are using our application for demo please use Firefox, web-app because it does not support in chrome since chrome do not allow web-cam access.
what will be the situation of the people if they are not allowed to travel around the city or across the world due to COVID-19 for the long time ? what will be the safest way to travel around the city or world? How can we automated the process through AI? How can our solution helps at this time and COVID-19 pandemic is over?
To get the solution of these question, we did some research to identify the challenges to developed the best solution.
fig: covid-19 in the world
As we see from the map, COVID-19 is now world epidemic problem. Novel corona virus has killed 775 893 people and infected over 21 989 366 people around the world wide during preparing this report. This number is growing everyday.
fig:Data according to WHO(2020 Aug-10)
The main problems occurs in the transportation during the pandemic are:
1. Contamination : As you may see on the news, corona virus outbreak is possible through public vehicles mostly. People are using the public vehicles without social distance and they are not using any safety equipment. Therefore to reduce contamination our application checks the safety measure.
2. Mask and Glove Safety Check : Gloves, protect the hands, gowns or aprons protect the skin and/or clothing, masks and respirators protect the mouth and nose, goggles protect the eyes, and face shields protect the entire face. Our system aware to use mask and gloves if they do not follow people are not allowed to enter into vehicles.
3. Contact Tracing : During this pandemic contact tracing is very important in order to find the contaminate people through public vehicles but after this system implementation it is much more easier to trace using their ID information, scan before enter into vehicles.
4. Security in Public Vehicles : In other time also security is one measure issue inside the public vehicles and difficult to trace the criminal however this application greatly reduce it because Personal ID is necessary to use public vehicles whose ID information are store in the database that makes tracing faster and easier.
What it does
Using the power of pre-trained machine learning models from open source, PasssengerCOVIDscan-AI is created as a full-scaled AI tool for Transportation. It can automate the process of security ,mask and gloves safety check for passenger.This tool of cutting edge technology can be used to reduce criminal activities and speed up tracing time for police in this critical time of the COVID-19 pandemic.
The PasssengerCOVIDscan-AI web-app includes 3 main AI components:
1. Passenger ID Scanner: For the safety purpose, only passenger with their Personal ID is allowed to enter into public vehicles. Hence, the web-app will use pre-trained data to scan the people ID and grant them access to the vehicle.
Fig: ID card Scanning
Fig: Scan Data information in the form of E-mail
2. Mask and Glove Safety Check:
Especially during this COVID-19 ourbreak, the web-app will ask the passenger whether the passenger is used mask and glove or not and have ID too. If yes, then the web-app use AWS pre-trained data to check the safety measures of the people to minimize any exposures to the disease. If the passenger passed all the secured check and safety, he/she can travel by vehicles.
Fig:Demo of mask scanning
How we built it
- Passenger ID Scanner: We developed this feature using the open-source python library Pyzbar. We have written the script in the JQuery which sends the snapshots from the live camera feed to the inference model at the back-end. It can read one-dimensional barcodes and QR codes present on the passenger ID badge. We implemented this feature to work with a snapshot of passenger ID . Link: https://pypi.org/project/pyzbar/
2.Mask and Glove Safety Check: We developed this feature using the open-source TensorFlow model for face mask detection. The backbone network only has 8 Convulation layers and the total model has only 24 layers with the location and classification layers counted. We have written the script in the JQuery which sends the snapshots from the live camera feed to the inference model at the backend. It works with live footage from any sort of cameras and detects people not wearing a face mask and glove. Link: https://github.com/AIZOOTech/FaceMaskDetection
Technical Requirements: The packages required for this project are as follows:
Challenges we ran into
This is the first time we all were working with such project and creating endpoints of the pre-trained TensorFlow model. we had a little problem of focus with certain cameras so we had to experiment with several webcams that we had available to find one that didn't require to focus.Due to poor web-cam it probably lack of resolution.
Accomplishments that we're proud of
We manage to finish the project in such a limited time of week in our free time from work. We still keep striving to submit on time while learning and developing at the same time. We are really satisfied and proud of our final product for the hackathon.
What we learned
Through this project, we learn to implement a complicated image-recognition deep learning models. We also learn the deploying the models in web-app. This project can’t be done without the efforts and collaboration from a team with such diverse backgrounds in technical skills.
What's next for PassengerCOVIDscan.AI
1) In coming days we would like to implement this ides if we get fund from the interested companies.
2) We will partner with research team of that company in order to collect the data.
3) We also plan to use Thermal Detector in the door of public vehicles to measure the temperature of the passenger if any unusual, alert is given to the near police station.
4) After the end of corona virus software is implement to reduce crime, harassment etc. in public vehicles i.e support police activities.