A UAV PHOTOGRAPHIC PATH PLANNING METHOD FOR HIGH-QUALITY RECONSTRUCTION OF CULTURAL HERITAGE

Abstract. The image-based reconstruction method can preserve geometric and textural information with relatively high accuracy, making it a suitable for digitally documenting cultural heritage. However, the quality of reconstructed model largely depends on captured images. Unmanned aerial vehicles (UAVs) equipped camera gimbal offer great convenience image acquisition in 3D reconstruction. ensuring safety, efficiency, full coverage is challenge. To address this, we propose UAV photographic path planning efficient automatic heritage scenes, based which high-quality realized. A priori proxy scene obtained advance utilized to (1) generate initial viewpoints subsequent optimization; (2) SDSM obstacle avoidance, signal analysis, sight occlusion judgment; (3) segment obtain planar regions sample representative points measuring reconstructability optimizing viewpoints. Our enables regular safe final paths heritage, outperforming both commercial software state-of-the-art methods real virtual scenes.