Recently, advanced and semi-automated equipment, laser-scanner and Unmanned Aerial Vehicles (UAVs), has been used for construction project. Especially, photogrammetry using image acquired from UAVs can make high accuracy data and need low cost compared with terrestrial and aerial Lidar. Unmanned Aerial vehicles (UAVs), aerial movable image sensing platform, have started to in popularly in lots of industries such as agriculture and environment assessment. However, in construction site, it is very hard to acquire high quality data from UAVs due to the difficulty in environmental conditions such as asymmetrical shapes, difficulty of setting control points, and post-processing. This study aims to propose the pre-processing methodology which is to compensate 2D image using image histogram for construction site. This method is for control a variety 3D building data as point-cloud using the pre-processing methods. Also, this research shows increased accuracy by comparing with and without pre-processing of source data getting from UAVS.