Advancing Low-Cost, Large-Scale Flow Visualization: A Case Study with Linear Camera Arrays in a Slow-Flow Wind Tunnel

We present a low-cost, scalable method for visualizing flow fields in the slow-flow turbulent atmospheric boundary layer wind tunnel at the University of New Hampshire. A linear array of three synchronized Raspberry Pi microcomputers, each connected to a camera module, captured time-resolved images at flow speeds of approximately 1–2 m/s. The system was configured with overlapping fields of view to form a continuous measurement corridor about two meters in length. This setup was deployed to record image sequences under a range of controlled flow conditions, using various flow tracers introduced to visualize the flow.

The recorded sequences were analyzed using motion estimation techniques, including particle tracking velocimetry and Farnebäck optical flow, to compute planar velocity fields. The estimated fields showed expected changes in flow speed and direction and were in general agreement with Pitot-static tube measurements, suggesting the system can provide accurate information about how velocity varies across space.

RaspiTrack offers a practical, low-cost alternative to traditional multi-camera flow visualization systems. Future work will focus on enhancing the system’s quantitative capabilities through algorithm refinement, improved calibration, and continued comparison with reference data. The system will be scaled to support nine or more synchronized cameras, enabling longer tracking regions and multi-view Lagrangian reconstruction.