The invention discloses a tunnel back-break detection method. The method includes the following steps that S1, according to center line data of a tunnel line, the thickness value and width value of a section are set, and tunnel section
point cloud automatic partitioning is carried out; S2, the partitioned section
point cloud is projected to the XOY plane, and a convex
hull extraction method is adopted for carrying out automatic extraction of a section
point cloud contour lines to obtain a measured section line; and S3, the tunnel design section is utilized, the distances between feature points of the tunnel design section and the measured section line and the locations of the feature points are calculated, and back-break automatic detection is carried out. Tunnel back-break detection is carried out by utilizing three-dimensional
laser-point
cloud data, and rapid partitioning of the section point cloud is achieved by adopting
radius searching and a rectangular partitioning
algorithm; on the basis of back-break area statistics of a polygonal intersection
algorithm, the method is suitable for back-break detection of various types of tunnels; and compared with a traditional total-
station section measurement mode, the detection efficiency is high, results are comprehensive and full and accurate, and the
measurement precision meets the back-break detection requirement.