Research on Surface Super Resolution for 3D optical scanning

DOI

Abstract

<p>Super resolution is a well-known technique in 2D image processing, and it has been an active research topic over the past few years. Compared to 2D image, 3D scanning data represents a more detailed geometric information of objects, which is often useful for reverse engineering to improve products processing flow, designed structure, etc. Generally, 3D scanning data needs to be highly accurate in reverse engineering and other applications. Due to the limitation of 3D scanning technique, point cloud data becomes inaccurate, and some details are lost. Based on classical super resolution theory in 2D image processing, we give a numerical evaluation method of surface super resolution for 3D optical scanning data by combining many similar scans. The contributions of our work are as follows: (1) Investigate the noise distribution of scanned data and simulated the scanned data; (2) Error evaluation on simulation of scanned data super resolution model;(3) Evaluation on super resolution effect for real scanning data. Extensive experimental results demonstrate the effectiveness of the super resolution method.</p>

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top