Wilderness Search
Wilderness search and rescue (WSAR) research focuses on developing efficient and robust methods for locating missing persons in challenging natural environments. Current efforts leverage advancements in computer vision, employing deep learning models (like EfficientDET) and multi-modal data fusion (combining visual and thermal imagery from drones) to analyze large datasets of aerial imagery, often enhanced by techniques like matched filtering and saliency detection to identify relevant features. These improvements aim to significantly reduce search times and improve the success rate of WSAR operations, impacting both the safety of missing persons and the efficiency of rescue efforts.
Papers
Confident Naturalness Explanation (CNE): A Framework to Explain and Assess Patterns Forming Naturalness
Ahmed Emam, Mohamed Farag, Ribana Roscher
Leveraging Activation Maximization and Generative Adversarial Training to Recognize and Explain Patterns in Natural Areas in Satellite Imagery
Ahmed Emam, Timo T. Stomberg, Ribana Roscher