Machine learning for humanitarian disaster relief efforts through employing rule-based verification on drone aerial imagery

The cutting-edge technologies can help to improve the operational response by facilitating evacuation for aged-care facilities and reducing the number of casualties during flood events. This research proposes a framework based on artificial intelligence machine learning technologies along with the application of aerial imagery through unmanned autonomous vehicles (UAV) technology. The UAV swarm will allow for the collection of images from the affected regions (age care facilities) in the Hawkesbury valley, providing the gathered data to the control centre. The drone-based collection of real-time data around the aged care facility centres during a flood event can be effectively analysed using various pre-processing and edge detection methods for feature extraction and labelling of the flood images. 

The UAV coverage and deployment of multiple UAV images are extremely useful for the analysis of road network infrastructure, bridges and other means of transportation in the near vicinity of the age care facilities in the Hawkesbury region. The machine learning methods are then applied to extracted images of flood for training the models. This will further facilitate the development and construction of the rescue networks for the age care facility providers in the Hawkesbury region to be used by the rescue teams. 

Knowledge contribution through this research study can potentially go a long way as the framework developed in this project can further be used for the analysis of infrastructure and planning disaster resilience. Additionally, the use of such integrated technologies (cloud computing, image processing, and artificial intelligence) for the collection of real-time information will assist in finding the appropriate routes to reach the disaster-effected areas with a potential improvement in the response time.