Reconstruction of Shapes using Machine Vision System Based on Focus Measurement: A Quantitative Analysis of Texture Influence Using 3D Printed Objects.

In the present scenario, the 3D reconstruction of a scene plays a vital role in many manufacturing industries. In this work, an attempt was made to develop a machine vision system under shape from focus technique for surface reconstruction with active illumination pattern projection. The accuracy of the reconstruction of the proposed method was investigated and compared with the traditional approach. It was found that the developed approach can provide enhanced accuracy owing to the improved texture of the samples by additional pattern illumination and removing the noise content by a dynamic window size median filter. Among all shapes considered, Hemisphere which has more height and curved feature performs better reconstruction. [ABSTRACT FROM AUTHOR]