Abstract: The complex wave structures generated by a ship significantly influence the wave-making resistance, which is a critical factor in optimizing vessel performance and hull design. Traditional measurement methods, such as wave height gauges, are limited by sparse measurement points, low spatial resolution, and potential interference with the wave surface, making them unsuitable for full-field wave surface measurements in complex wave systems. To address these limitations, a non-contact measurement method for ship wave surfaces based on multi-view stereoscopic vision is presented. The proposed scheme includes: applying neural networks for spatial calibration and the three-dimensional reconstruction of feature points, using an ant colony particle tracking algorithm for perspective matching, and conducting principal component analysis for still water baseline correction. Experimental results demonstrate that the proposed method can realize the wide-area free-surface measurement of the ship wave and effectively capture typical wave characteristics at various ship speeds. Compared to traditional measurement, the stereoscopic vision approach offers enhanced flexibility and suitability for real-time measurement of dynamic wave fields over large areas.