Unsharp Mask Guided Filtering for Acoustic Point Cloud of Water-Conveyance Tunnel

The inner-surface damage of water conveyance tunnels is the main hidden danger that threatens their safety and leads to serious accidents. method based on principle acoustic reflection means inspecting water-conveyance tunnels. However, affected by tunnel environment equipment noise, obtained point cloud model inevitably suffers from which can produce erroneous results. Therefore, we proposed a novel filtering method, called unsharp-mask-guided for 3D cloud, reduce impact noise fuses ideas guided unsharp masking technique extends them considering position point. In addition, edge-aware weighting mean also used retain edge features while smoothing points. experimental results show our obtain impressive better performance in both simulated than many state-of-the-art methods.