The main purpose behind our project is to unite lost people with their families, especially during the time of calamities when the percentage of people separated from their families is comparatively high.
During the time of disasters/ calamities unfortunately a lot of people are separated from their families and it takes a lot of time and legal procedures before they meet again with their families. Also during such conditions, there is an extensive coverage of such events with numerous cameras by various agencies, thereby giving us a huge amount of data to exploit into. This data could be used to detect the missing people and could be further processed to match them with the missing photo present in the database.
- The reporter uploads short videos of the calamity recorded by him on the website with all the relevant details pertaining to the location of the video.
- The videos get added to the folder containing all the previously uploaded videos. At the same time, each video is fragmented into frames and all the frames belonging to each video is saved to its corresponding folder.
- For each video, all the frames are checked for a face. All the frames which do not contain a face are discarded.
- Families of missing people file a report with the police, who in turn uploads the photograph of the missing person on the website with all the relevant details.
- For each missing report, the photo is matched with each frame of each video using facial recognition algorithms.
- A match with a frame confirms the presence of the missing person in the area in which the video was taken. Such a match triggers a notification to the police with the all the details of the uploaded video including the details of the reporter and the location of the video.
DJANGO, Bootstrap
Python, Javascript, CSS, HTML
Facial recognition algorithms
Install the following libraries
- Django
- face-recognition
- imutils
- matplotlib
- numpy
- opencv-python
- glob
- python manage.py migrate
- python manage.py makemigrations
- python manage.py migrate
- python manage.py runserver
- On the home page, click on the "CLICK HERE TO UPLOAD THE VIDEO" link.
- This will redirect you to a page where you can upload the video from the "foruploadingpurposes" folder. Type your location too and click Upload.
- After that, sign yourself up and log in.
- This will redirect you to the homepage, where you need to click on the "POLICE! UPLOAD DETAILS OF A LOST KID" link.
- This will redirect you to a page where you can upload the image from the "foruploadingpurposes" folder. Type other details too and click "Find the missing person".
- This will redirect you to the result page with the message that the person was found in the location that you typed because the person in the image is there in the video.