Nearshore space-time ocean wave observation using low-cost video cameras

A low-cost stereo vision technique is presented to estimate the ocean wave field in space and time. The method costs only about 1% of traditional instruments for wave measurements and is even cheaper and easier to implement than typical stereo video techniques. Despite the system providing lower-resolution data than higher-grade stereo video systems with advanced optics and synchronisation tools, the low-cost video synchronisation scheme does not require dedicated acquisition software, triggers, and cables, making the system completely portable. The system's performance was validated using a co-located pressure gauge, enabling wave measurements during experimental campaigns conducted in shallow water and the surf zone. The wave data obtained using the proposed method with two action cameras showed very small deviations in significant wave height measurements and effectively identified the dominant frequency in the wave field. Although this system is currently designed for a fixed setup, it can be adapted for use on mobile platforms such as ships, drones, or other moving systems, provided that appropriate motion correction is applied. The results indicate a higher significant wave height when considering spatial wave fields than in the temporal point elevation series. The low-cost stereo video method is a promising approach for acquiring spatiotemporal wave data for specific wave-related engineering problems. • A low-cost stereo vision technique is presented to estimate the ocean wave field in space and time. • The method costs only about 1% of traditional instruments for wave measurements. • The video synchronisation does not require dedicated acquisition software, triggers, and cables, making the system portable. • The performance of the system has been validated with a co-located pressure gauge, allowing the measurement of the waves. • The experiments evidenced that surface displacements close to the surf zone tend not to behave as a strictly Gaussian process.