A method for identifying circumferential joints and detecting inter-ring misalignment in metro shield tunnels based on point cloud intensity features and Hough line detection

Efficient and accurate detection of inter-ring misalignment in metro shield tunnels is essential to structural safety. Existing detection methods based on mobile laser scanning point clouds rely on bolt holes or grayscale images to assist circumferential joint identification and involve cumbersome processing procedures. To address these limitations, a method for identifying circumferential joints and calculating inter-ring misalignment is proposed based on the reflectance intensity features of three-dimensional laser point clouds and Hough line detection. In this method, the three-dimensional point cloud was unwrapped onto a two-dimensional plane; circumferential joints were directly located in the point cloud by combining reflectance intensity filtering with the Hough transform; noise from ancillary facilities was suppressed using a two-stage ellipse-fitting strategy combining Random Sample Consensus with least squares-fitting; and inter-ring misalignment was calculated through cross-section matching. The method was tested and validated using point cloud data from a metro shield tunnel. The results show that the method yields a mean absolute error of 3.32 mm and a root-mean-square error of 4.69 mm in circumferential joint localization. For 60 sets of misalignment measurements taken at fixed angular positions, the mean absolute error, root-mean-square error, and 95th-percentile error are 1.028 mm, 1.296 mm, and 2.484 mm, respectively. The verification results for the maximum inter-ring misalignment that the maximum absolute error is 3.3 mm.