Abstract: In deformation monitoring of subway tunnels, 3D laser scanning technology could obtain the high-precision point cloud data in a simple, fast and real-time way, and use the data for tunnel engineering structure analysis. A solution with high-precision and high computational performance is proposed for tunnel ellipticity detection according to the massive point cloud data. Firstly, the tunnel point cloud data is imported into the database for storage management and index construction; then, through parallel computation, rapid process actions including automatic denoising, ellipse fitting and parameter extraction for section point cloud is conducted. Experiment results show that the proposed detection solution has the advantages of high-precision and high computational performance for the analysis of massive tunnel point cloud.