At present,time-of-flight (ToF) technology is widely used in 3D imaging;however the method of 3D point cloud target extraction is rarely.Therefore,we present a 3D point cloud complex target extraction method based on point cloud location information and intensity information.Firstly,the edge extraction and morphological processing are used to locate the target position in the intensity information,so that the target can be identified in the intensity image.Then the intensity information is correlated with the position information,and the region of the target is extracted from the 3D point cloud.Finally,by establishing a K-neighboring,and by improving the selection method of the neighborhood size,the Gaussian mean replaces the fixed artificially selected neighborhood.The experimental results show that this method can effectively filter outliers and retain the position and intensity information of the real object point cloud data,and has a high computing speed.