This paper presents a novel processing architecture for unmanned aerial vehicles (UAVs), designed to support small search-and-rescue teams operating in post-disaster environments. The proposed system consists of two main components. The primary component is dedicated to vehicle and person detection and tracking. It utilizes a Raspberry Pi 5 paired with a Coral TPU Accelerator to process input from the UAV's main camera, leveraging a YOLOv11n neural network and the KCF tracking algorithm. This capability is essential for following the rescue team's vehicle and locating survivors. When a network connection is available, this subsystem transmits captured images to a remote server for further analysis. The secondary component is a peripheral vision subsystem, powered by a Raspberry Pi Zero 2 W. It processes input from four peripheral cameras connected to a Luxonis OAK-FFC-4P board. It is tasked with identifying critical conditions and communicating the findings to the main processing unit, which in turn alerts the operator. Once appropriate instructions are received, the UAV is redirected to the specified location to perform a focused search for survivors.

The increasing movement of personnel and vehicles in airport taxiway areas raises the risk of ground accidents involving aircraft. This paper presents a real-time, camera-based monitoring system designed to prevent such incidents by detecting personnel in proximity to aircraft and issuing timely alerts. The proposed solution integrates advanced computer vision techniques with a client-server architecture, where a lightweight edge device transmits captured frames to a remote server for intensive processing. The system (i) detects aircraft, (ii) detects personnel, (iii) calculates the distance between each aircraft and the camera, (iv) calculates the distance between each detected person and the camera, and (v) calculates the 3D distance between aircraft and personnel. In case a person is identified within a predefined safety radius around an aircraft, an audio alarm is immediately triggered to warn ground staff. Experimental results demonstrate that the system achieves an accuracy of$\mathbf{9 5 \%}$in calculating the aircraft-to-person distance. By combining machine learning, real-time processing, and automated hazard evaluation, the system reduces reliance on human observation, minimizes the risk of human error, and enhances overall airport ground safety. The complete pipeline achieves a frame processing rate of approximately 16 frames per second, ensuring timely and continuous monitoring of fast-moving targets.

The present work focuses on the study of particle deposition in segments of the cardiovascular system. In particular, the geometry of an iliac bifurcation is reconstructed from medical imaging data and the flow fields of both blood and particles are obtained using Computational Fluid Particle Dynamics techniques. Particle convection, diffusion and inertia are taken into account in the simulations. The numerical experiments indicate that diffusion dominates deposition and only 10% of the injected particles deposit in the bifurcation. Both blood flow field and the characteristics of the patient specific geometry influence the particle deposition sites. Overall, the proposed methodology could become a useful tool for the design and optimization of biomedical applications.