Utilising gamification approaches to derive crowd pattern/crowd context from aerial images of major events

Dissertation submitted in partial fulfilment of the requirements for the degree of Master of Science in Geospatial TechnologiesA large number of casualties occur during emergencies in highly-crowded public spaces of major events like annual anniversaries, religious festivals, big parties and football matches due to stampedes. It has been often observed that poor resource management is one of the key areas that could be improved to solve this problem. In this study,a geo-game-based approach has been adopted to alert responsible authorities of highly crowded regions as an early warning system and also provide them with optimal dispersal routes. In the Android-based game that was developed for this study, the players could draw polygons on real-time imageries of the area under study obtained from unmanned aerial vehicles and classify them into categories based on how crowded the region is. This data is submitted to a web server which is processed to find suitable least-cost routes by which the people in the crowded regions can be brought to safety in case of an emergency. The spatial distribution of people could be forwarded to appropriate authorities in-charge of the administration highlighting extremely crowded regions which need their attention thereby prompting redirection of security personnel. Additionally, the calculated dispersal routes could be used by them as suggestions to avoid a stampede and ensure safety in case the situation turns worse