Enhancing People Localization in Drone Imagery for Better Crowd Management by Utilizing Every Pixel in High-Resolution Images

Accurate people localisation using drones is crucial for effective crowd management, not only during massive events and public gatherings but also for monitoring daily urban crowd flow. Traditional methods for localising tiny objects using high-resolution drone imagery often face limitations in precision and efficiency, primarily due to constraints on image scaling and the use of sliding-window techniques. To address these challenges, a novel approach dedicated to point-oriented object localisation is proposed. Along with this approach, the Pixel Distill module is introduced to enhance the processing of high-definition images by extracting spatial information from individual pixels at once. Additionally, a new dataset, UP-COUNT, tailored for contemporary drone applications, is shared. It addresses a wide range of challenges in drone imagery, including simultaneous camera and object motion during image acquisition, thereby advancing the capabilities of crowd management applications. A comprehensive evaluation of our approach on the proposed dataset and the commonly used DroneCrowd dataset shows that it outperforms current approaches, highlighting its efficacy in drone-based crowd object localisation tasks under standard acquisition conditions. These improvements significantly enhance the algorithm’s applicability in real-world scenarios, enabling more reliable localisation and counting of individuals in dynamic, well-lit environments.