Integration of Deep Learning Object Detection Techniques and Drone Technology for Disaster Monitoring and Detection

The scale and frequency of natural disasters have increased in the past few years due to environmental change and worldwide warming. Natural disasters occur on a daily basis throughout the world, and represent a major factor affecting the environment and the lives of the human population. These catastrophic events can result in immeasurable human and material losses, as well as devastating economic, social and environmental consequences, often leaving communities in ruins. Even if we can't fully control or predict them, we can harness the power of technology and innovation, through advanced drone technology and artificial intelligence, to mitigate the impact of these disasters. The use of drones incorporating deep learning techniques offers a cost-efficient remedy for proactive disaster detection and real-time monitoring. In this article, we take a detailed look at recent advances in the realm of object detection by deep learning, including models such as R-CNN, YOLO and their variants, focusing on their potential applications in disaster monitoring. Experiments carried out demonstrate promising outcomes across various metrics, rendering it a valuable asset for disaster monitoring and detection. Thus, comparison was made between these two models, and each with its own properties. Each of the models performed well in the disaster recognition application, with YOLOv8 being the optimum model for real-time use due to its combination of speed and accuracy. The other model, Faster R-CNN, also performed well in a variety of use cases.