MiPIS (Missing Person Identification System) is an WebApp which can be used for tracking of missing people based on inputs from users.
The backend is controlled by Django and Python. And uses MySQL for the database to store data. The face similiarity detection is backed by Face++ API. The SMS service used to notify the person is provided by Vonage API.The Face similiarity utilised is K-Means clustering.
The Front End is made through HTML/CSS and JS.The installation is simple as described as follows.
- Finds out a person based on image submited by the user and notifies the concerned authorities.
- Uses Face Recognition as well as Manual Approval before confirming a find claim by a user.
- User is notified through SMS about the whereabouts of the missing person through SMS message.(currently on trial tier so only works on whitelisted numbers)
- A leaderboard of users based on persons correctly found by them!
- Dockerization(in process)
- User side of the Website can be integrated to a mobile App.
- MySQL needs to be configured with local setting in settings.py
- A Media directory needs to be created if not cloned dirctory. Should be at same level as templates.
- Other libraries used are mentioned in
requirements.txt
- Add the
Server
file to your path. - It is advised to create a virtual environment in python or use
source MyEnv/bin/activate
in the parent cloned directory the activate My Virtual Environment. - Now install the required libraries by typing
pip install -r requirements.txt
to install them at once. - Ensure that MySQL client is setup and running also the settings are properly configured in
settings.py
. If Using Arch linux install mariadb and start it by typingsystemctl start mariadb.service
.And create a database called MiPIS byCREATE DATABASE MiPIS;
. - Now at the same directory as
manage.py
run the server through Gunicorn.gunicorn Server.wsgi
orpython manage.py runserver
. Whatever works out fine. - Now you can open the WebApp in your LocalHost.
The Webapp is also hosted here MiPIS. But due to issues with free tier of PythonAnywhere the face API Times out and gives undesirable issues so local version is preferred.
Thank You For Using Mipis.