ABSTRACT

This study investigates the feasibility of using binocular vision technology for reconstructing wave shapes in a recirculating water flume with a strong three-dimensional bow wave. A custom-built binocular vision system is utilized to reconstruct the free surface of both unbroken and breaking bow waves. The technical methodology includes calibration methods, epipolar rectification, high dynamic range imaging (HDRI) techniques, and cross-correlation matching. The impact of HDRI and matching block size on the feasibility and accuracy of binocular reconstruction is analyzed. The findings indicate that HDRI significantly enhances the reconstruction reliability of non-breaking wave surfaces, while its effect on broken areas is minimal. The smaller matching block sizes improve the reconstruction reliability in broken areas but exhibit a non-monotonic effect on non-breaking areas. These study proves that the present binocular system can accurately reconstruct the three-dimensional shapes of bow waves, irrespective of surface breaking.

INTRODUCTION

Understanding the three-dimensional shape of ship bow waves is essential for advancements in maritime engineering and hydrodynamics. These waves, especially at high speeds, create complex flow phenomena like wave breaking and air entrainment, significantly affecting a ship's resistance and maneuverability. These phenomena also impact fuel efficiency, operational safety, and environmental footprint. Specifically, accurately predicting and understanding bow wave patterns can lead to more efficient ship designs, reduced fuel consumption, and lower emissions (Ntouras et al. 2022). The ability to model and predict bow wave behavior is crucial for optimizing ship performance under various conditions, enhancing overall maritime safety and efficiency (Dong et al. 2022; Noblesse et al. 2013). Moreover, understanding bow wave dynamics helps mitigate environmental impacts, as efficient ship designs lead to lower emissions and less disturbance to marine ecosystems (Gabel et al. 2017).

Physical experiments are crucial for studying bow wave phenomena. Traditional methods include wave probes, high-speed cameras, ultrasonic sensors, etc. The most commonly used capacitance or resistance wave gauges provide precise measurements of wave height (Olivieri et al. 2003), but can disturb the flow and offer limited point measurements to capture full 3D shapes. High-speed cameras capture detailed wave surfaces, allowing visualization of wave breaking and air entrainment (Waniewski et al. 2002). Video-based methods also capture wave dynamics by recording wave motion and analyzing the images to determine wave height and other parameters (Kim et al. 2016). The extracting quantitative data from the single camera's or video's images are always two-dimensional. Ultrasonic sensors offer a non-invasive option, measuring wave elevation accurately without disturbing the water surface, useful in dynamic marine environments. Some laser-based techniques like stereo-correlation and stereo-refraction could also provide the wave shape data (Gomit et al. 2014). However, the ultrasonic sensors and the laser-based techniques tend to fail when the wave surface undergoes breaking. Overall, these aforementioned methods struggle to capture the complete three-dimensional shape of wave surfaces, especially in dynamic and turbulent conditions (Wang et al. 2020). Therefore, advanced techniques are needed to provide comprehensive 3D data of bow waves.

This content is only available via PDF.
You can access this article if you purchase or spend a download.