Louie Lu, PhD 

Close-range Photogrammetry

The goal of close-range photogrammetry is generation of a hihg accuracy 3D model. This project focuses on removel the traditional coded target from images, which provide better flexibility to the traditional photogrammetry. In order to avoid the use of target, the SURF algorithm is applied to generate point correspondences. However, this correspondence contains quite amount of missmatches. The proposed approach use a machine learning approach to detect and reomove outliers form point correspondence, which provide a uncontaminated point correspondences for calculating relative orientation.

Just a sample of my work. To see more or discuss possible work >>