High dynamic depth range for 3D image capturing system

Yi-Pai Huang; Po-Yuan Hsieh; Yong-Ren Su; Han-Ping D. Shieh

doi:10.1117/12.2054432

Translator Disclaimer

5 June 2014 High dynamic depth range for 3D image capturing system

Yi-Pai Huang, Po-Yuan Hsieh, Yong-Ren Su, Han-Ping D. Shieh

Author Affiliations +

Proceedings Volume 9117, Three-Dimensional Imaging, Visualization, and Display 2014; 91170F (2014) https://doi.org/10.1117/12.2054432
Event: SPIE Sensing Technology + Applications, 2014, Baltimore, Maryland, United States

Abstract

To detect the 3D depth information of objects in a deep scene is not so easy due to the limited depth of field (DoF) of cameras. In this paper, we proposed a 3D depth map capturing system with high dynamic depth range (HDDR). Unlike conventional extending depth of field (EDoF) method, the HDDR method will not deteriorate the image quality. By imitating an active tunable m × n lens array focusing on a sequential of imaging planes, each object in the scene would be clearly captured by at least three elemental lenses. We estimated the elemental depth maps by using the method depth from disparity individually, and then fused them into one all-in-focus depth map. Comparing with the conventional 3D cameras, the working range of HDDR system with 3x3 camera array can be extend from 90cm to 165cm.

Citation Download Citation

Yi-Pai Huang, Po-Yuan Hsieh, Yong-Ren Su, and Han-Ping D. Shieh "High dynamic depth range for 3D image capturing system", Proc. SPIE 9117, Three-Dimensional Imaging, Visualization, and Display 2014, 91170F (5 June 2014); https://doi.org/10.1117/12.2054432

ACCESS THE FULL ARTICLE