Adaptive LiDAR scanning based on RGB information

Accurate understanding of scenes through vision can play a critical role in construction automation technologies, particularly in navigating the challenges of occlusions and object stacking. However, methods are needed to adaptively gather essential data for precise detection, bypassing unnecessary data storage and computational expenses, to strike a balance between capture robustness and efficiency. This study presents an adaptive reality capture method to enhance object segmentation and detection in complex geometries. A two-tiered approach is applied, using RGB data to pinpoint uncertain areas, followed by LiDAR scans for detailed data acquisition. This method improves detection accuracy and efficiency by focusing on critical regions, reducing computational demands and scan time. The results highlight the potential of adaptive methods to achieve remarkable detection outcomes in construction environments with minimal computational costs, showcasing efficient data management. Additionally, this research suggests a promising direction for future advancements in intelligent, adaptive sensing technologies within construction automation. •An adaptive LiDAR scanning workflow is proposed using RGB data for preliminary scene uncertainty estimation.•Shannon information entropy formulation is used to quantify the required LiDAR points to increase the detection confidence.•Comprehensive metrics are used to evaluate the benefits of adaptive LiDAR scanning versus uniform scanning.•It presents a solution for quality scanning without significantly increasing the number of point cloud data.