In view of the problem that the hugeness of the point cloud data of laser measurement normally leads to the slow running speed of computer and the waste of resources, and based on the defect of angle deviation method, an angle distance method, combining the point curvature, angle and distance together, has been proposed to simplify the laser measurement data. The proposed method changes the point curvature and the angle threshold into a function that can adjust the data size according to the curvature of the data points. Meanwhile, in combination with the minimum distance method, it sets the distance threshold, with the data points larger than the threshold preserved. The experimental results show that this method can effectively retain the points with large curvature value without breakpoints, which verifies the fact that accuracy and reduction data are better than angle deviation method in reduction effect.