A light field measurement system through PSF estimation by a morphology-based method

Abstract Light field imaging technology can obtain three-dimensional (3D) information of a test surface in single exposure. Traditional light reconstruction algorithms not only take long time to trace back the original image, but also require exact parameters system, such as position and posture microlens array (MLA), which will cause errors reconstructed image if these cannot be precisely obtained. This paper proposes algorithm for based on point spread function (PSF), does prior knowledge system. The accurate PSF derivation process system is presented, modeling simulation were conducted relationship between spatial distribution characteristics A morphology-based method proposed analyze overlapping area subimages images identify location MLA used thereafter accurately refocus imaging. built verify algorithm’s effectiveness. Experimental results show that measurement accuracy increased over 41.0% compared with traditional by measuring step standard. improved through microstructure peak-to-valley value 25.4% root mean square 23.5% improvement. further validates effectively improve refocusing efficiency superiority without provides new solution fast 3D field.