Efficient Geometric Quality Inspection of Large-Scale Ship Components via Nap-Of-The-Object Photogrammetry

Abstract. Ship components in the shipbuilding industry are often large in size and structurally complex, making it difficult for conventional quality inspection techniques to simultaneously satisfy efficiency and accuracy requirements. To address this issue, this paper proposes a geometric quality inspection method for ship components based on nap-of-the-object photogrammetry. First, a UAV flight path is planned according to the CAD design model of the target component, and a micro-UAV is used to acquire high-resolution images. High-precision 3D reconstruction is then performed by integrating exterior orientation parameters such as heading, attitude, and gimbal angle. Finally, geometric primitive fitting between the 3D reconstruction model and the CAD design model is conducted to evaluate geometric indicators including perpendicularity, parallelism, coplanarity, and detailed deviations. A segmented ship component was selected as the experimental subject. The measurement results obtained from the 3D reconstruction model were compared with the corresponding data from the CAD design model. The results demonstrate good perpendicularity, parallelism, and coplanarity, with 90% of the measurement deviations within ±5 mm, and a maximum error of 1.46 cm, confirming the feasibility of the proposed method for geometric quality inspection of ship components. As a non-contact, high-precision, and high-efficiency measurement technique, nap-of-the-object photogrammetry can significantly enhance production efficiency and product quality in the shipbuilding industry, thereby promoting its modernization.