Frontend-repo(https://github.com/codelover4655/eng_frontend)
Backend-repo(https://github.com/codelover4655/eng_backend)
Find Missing People is a web application that helps simplify the issue of locating missing people, You can file complaints for missing people or if you found someone in stray conditions or if someone got lost from their loved ones. The application uses face recognization to match profiles for the Missing people and Found people and notifies both users whenever there is a match. Example - User A lodged a complaint about a missing person, and another User B founds a person and files a request, then if there is a match found then, both of them would be notified.
- Matching Missing Person Profile To Found Person Profile via Face Recognization Azure Api
- Email Notificaton when got a match
- Async Task scheduling via Celery on Backend Server For Better Performance and Less Waiting Time
- Fuzzy Search For Searching in List of Profiles for Better Filtering
- 👤 User Authenticaion
- Authentication also Supports Google Login
- [Presentation]
Deployed Website :( https://629392f60b696659d6578353--missingpeople.netlify.app)
- Frontend Server is Deployed to Netlify
- Backend Server is Deployed to Heroku (https://proper-backend1.herokuapp.com)
- PostGres Database on Heroku
- Celery On Heroku
- Media Files On cloudinary (https://cloudinary.com/)
- short video demostration to install this project on local setup https://drive.google.com/file/d/15bpbQhZyF_zKradEX_FgcIYMUFXR5KAA/view?usp=sharing
- To Run Project only need to Run Frontend () Backend Is already attched to it
- React, Bootstrap For Frontend Server
- Django Rest FrameWork For Backend Server
- Celery For Running Async Tasks On backend Server
- Azure Face Detection Api for Facedetection (https://docs.microsoft.com/en-us/rest/api/faceapi/face/detect-with-stream)
- Azure Face Verifiaction Api for Face verification (https://docs.microsoft.com/en-us/rest/api/faceapi/face/verify-face-to-face)
- Mapbox For embedding Maps (https://www.mapbox.com/)
- gecoding and reverse geocoding for Api's for fecthing Location from coordinate
- Auto Suggest Mapbox Api for Loaction Search Bar
- Smtp For Email Notifiaction From Backend Server(Django Rest Framework)
- OAUT 2.0 For Google Login
- Fuse.js For Fuzzy Search
- React Toast for alearts
- Cloudinary For Storing Media Files
- Used Sqlite for Local setup and Postgres for Deployment
- This web-application has varying permmisions for Visiting Diffrent Section of Website
- To see Home-Page Section , Missing People List Section , Found People List Section ANd Visting Any Full Profie Card User Dont need to Sign-in/Sign-up On this Website These Sections Are Open To All .
- For Filing any complain of Missing Or Found You Need to Login/Register in Web-Application as an User.Quick Login is available via Google Login.
- In Missing People List Section You can find every Missing Complain Which is Registered On my Application And Not Yet Taken Back. So Here You See Some Profie Card Which Has a Full detials Button Which when clicked Will Show You The Full Profie Regarding that card. The Found People List section is also designed in the same way.
- On Missing People List Section and Found People List you also get an Search bar where you can search profile via Name or father Name or Mother Name or Age. This is a Fuzzy Search so No need to be very confident with your spellings .
-
Filing Complain(Missing/Found)
- When Head To File complain section either Missing or Found You Have to Login in Webapp as a user. Then fill for with the details of whom you are fill This Form two thing to keep in mind first is enter your contact number in valid format(first writing country code then give space and write yourrest of the number without country code) and second thing is Upload clear and face detectable image . in either of the case you will get alert and have to fill complain again.
-
My Complain Section
- When head over to mycomplain section(only visible after login) you will see all missing and found complains registerd by you in seprated boxes there you got two buttons for every profie first is "CHECK STATUS" button where you can see what's going with your request means here you see all the matching profiles.
-
Mathcing Profiles
- Matching profiles means if you file a Missing Complain and someone else files a Found coplain or vise-versa and face both both the profils are similer (similarity is judged by Azure Face varifiaction Api) Then these Profies are categorized as matching profies.
- Another Profile control you get in My Profie Section is to delete Profie.
-
What Happens When Got a match
- Both The User involved in match got mail from website-email(missingperson4655@gmail.com) There they got link of Profile Of whom They Got Match
- As well as in my complain section they can check status where they can se those matching profiles
First Setup Backend such that when you start frontend it give full functionality.
- For Running this application with full funtionality On your Local Pc You to setup both backend-server(https://github.com/codelover4655/eng_backend) and frontend-server(https://github.com/codelover4655/eng_frontend)
-
STEPS TO SETUP FRONTEND SERVER
Pre-Requisites: You must have Nodejs installed locally on your system
-
Clone the Repository
git clone https://github.com/codelover4655/eng_frontend.git
or directly download the zip file. -
Move to the directry where this cloned file exist(if cloned in C folder directly so run commed in cmd
cd eng_frontend
) -
Install the Dependencies
npm install
-
Starting the App
npm start
ornode server.js
-
Open
http://localhost:3000/
to view the App -
be ensure this port is free for host
STEPS TO SETUP BACKEND SERVER
-
Clone the Repository
https://github.com/codelover4655/eng_backend.git
or directly download the zip file. -
install python on your pc if not present previously
-
Install virtual enviourment for python as there may be some dependencies which OS not give access to install on system directly
-
Process to install and create virtual envirment varies for diffrent operterating systems (supporting docs-https://docs.python.org/3/library/venv.html)
-
after activating virtual environment (if required)
-
Move to directry where you cloned backend files. root directory of app (
cd eng_backend
if cloned directly in C folder) -
Run
pip install -r requirements.txt
for installing all the dependencies -
after installing dependencies Run
python manage.py runserver 8000
-
Port must be 8000 because in frontend side it is linked with port 8000
-
Now Django server is running on local server
-
Now Run Celery Server which support Async Request Handling to Run celery Server
-
First install rabbitmq on your pc which servers as broker for celery(https://www.rabbitmq.com/download.html)
-
Run
celery -A eng_backend worker --pool=solo -l info
From the root directory of App -
After Both Servers Started running Backend starts giving Response to all Api Request.
- The Project has various rooms for improvements from scaling Application to adding new Features like As of Now i reduced the waiting time of user by using Async Task sechdulding like When User Files a missing complian i Run a Async task in backend to Search Whole database Foe similer Faces and send a Email notification when Found a match by this user may not have to wait for conformation but searching whole database is a time consuming task so This is a site of imporvement in my project.
😃 Suggestions and Contibutions are Welcomed! Feel free to Reach out to me! Tushar Singhal