The project was to develop an application that could count the number of people present at the site of a disaster in order to provide them better and faster support. This project was divided into two phases:
- Crowd counting using machine learning
- Mobile Application for displaying results
Need for the project
Disaster strikes at the most unexpected times. Floods, fire, earthquakes, etc. can lead to people being trapped in large masses. A disaster in a heavily populated region would mean that relief forces will have to dispatch an unknown amount of food, medication, clothes and other relief items. At Kumbh Mela or other mass gatherings prone to disasters, if we can estimate how many lives might be in danger in case of a disaster, we can arrange relief camps for the crowd at a safe place. The disaster-prone area can be tackled well.
- Quick help to people stuck onsite
- Better management
- Sufficient safety supplies for all people
- Concerned authorities can be informed about the headcount immediately
The images were gathered from all the available sources like media and social networking websites. We used image processing to smoothen the input image. Further, machine learning algorithm is used to count the number of people present in that image. The count is then generated as the output on the mobile application. This count was then sent to the concerned authorities to take further action.
We successfully built an app that could estimate the number of people in the crowd and inform the concerned authorities who could act accordingly.