Pipeline Defect Detection Based on Matrix-Array Solid-State LiDAR

To accurately detect surface defects in pipelines and obtain the depth and size information of the defect locations, this paper proposes a 3D point cloud detection method based on a matrix-array solid-state LiDAR. First, the collected point clouds undergo statistical filtering and voxel grid downsampling to remove noise and redundant data. Then, the Random Sample Consensus (RANSAC) algorithm is used to fit the standard plane. Next, a coarse registration is performed using the Sample Consensus (SAC) algorithm combined with the Fast Point Feature Histogram (FPFH), followed by fine registration with the Adaptive Iterative Closest Point (ICP) algorithm to achieve high-precision alignment between the defect point cloud and the reference point cloud. Finally, a region growing based clustering and segmentation method is used to extract the defect point cloud, enabling the quantification and visualization of the defect location. Experimental results demonstrate that this method can accurately extract pipeline defects and measure the depth and dimensions of the defect locations.