The Impact of Solar Angle and Cloud Shadows on 3D Reconstruction of Rolling Stock Cargo

Meeting the relentless demand for more efficient air cargo transportation is of paramount importance for commercial needs and military missions. This study describes an experiment to test an innovative approach that harnesses cutting-edge stereoscopic vision technology to create 3D point clouds of rolling stock cargo across varying solar angles and cloud shadow conditions. Virtual cargo point clouds are generated by calibrating and systematically organizing the depth and location points from an RGB-D camera and then reprojecting them in a virtual environment. Measurement accuracy was rigorously tested across six camera positions in various combinations of weather conditions against physical ground truth measurements. The high accuracy of such systems offers detailed, real-time insights into optimized cargo loading under unpredictable outdoor conditions. The findings, informed by both quantitative and qualitative analyses, reveal the impact of solar position, cloud coverage, and camera placement on the completeness of the point clouds, quality of the depth points, and accuracy of the measurements. This study provides insights to push the boundaries of cargo logistics possibilities in challenging outdoor environments.