Bridge pier is a building that supports bridge span structure and transmits dead load and live load to foundation. As the main supporting structure of the whole bridge, the perpendicularity of the pier is particularly important. Among them, for the pier with high height, the influence of perpendicularity deviation on the quality of pier is more obvious. In this paper, an automatic detection method of pier perpendicularity is proposed. The 3D laser point cloud is used to obtain the spatial attitude of the pier, and the point cloud of the reliable feature surface of the pier is extracted by combining the automatic recognition algorithm. According to the principle of statistical correlation, the characteristic surface data of piers are processed to get the actual center line and section of piers. At last, we update the BIM model of the designed pier with the center line and section and analyze the 3D deviation between the updated BIM model of the pier and the point cloud. We apply this method to the verticality detection of several bridges. The results show that the method is reliable and robust.