| dc.description.abstract | When out on a search and rescue mission, it is important to have tools that can easily keep track of the situation that is being handled. Autonomous drones have the ability to quickly collect a batch of images of the scene and its surroundings in order to provide emergency responders with an overview of what they are dealing with. These images are also used to identify hazardous anomalies such as tiny cracks on collapsed buildings. In many cases, however, identifying the exact location of these anomalies may be too difficult, especially when the anomaly is relatively minuscule in size when compared to the structure that it inhabits. The conducted research focuses on developing a system which search and rescue teams may utilize in order to extract the exact coordinates of any point found on an image taken by a drone. In order to do so, a series of images containing the scene of the area of interest is taken from a high altitude. Once that is completed, the images are loaded onto an application called Agisoft Metashape, where the images are combined in order to create a 3-Dimensional model of the location. Finally, the Image Coordinate Point Extraction algorithm, which was created using Metashape’s Python API, is run. The algorithm takes in an image as an input, presents it to the user, and asks the user to click a point on the image to extract its exact coordinate. In the case of this study, the entire process was tested on data from the Surfside Condominium building collapse that occurred in the summer of 2021. | |
