Rapid and Inexpensive Reconstruction of 3D Structures for Micro-Objects Using Common Optical Microscopy

TL;DR

This paper presents a quick, cost-effective method for reconstructing 3D surfaces of micro-objects using standard optical microscopy, image stacking, and open-source software, achieving high accuracy in under 10 minutes.

Contribution

It introduces a simple, inexpensive technique for 3D micro-object surface reconstruction using common microscopy and open-source tools, without specialized equipment.

Findings

Reconstruction accuracy of ~0.01 for 100 um features

Image acquisition time is approximately 10 minutes

Method successfully applied to various micro-objects

Abstract

A simple method of constructing the 3D surface of non-transparent micro-objects by extending the depth-of-field on the whole attainable surface is presented. The series of images of a sample are recorded by the sequential movement of the sample with respect to the microscope focus. The portions of the surface of the sample appear in focus in the different images in the series. The indexed series of the in-focus portions of the sample surface is combined in one sharp 2D image and interpolated into the 3D surface representing the surface of an original micro-object. For an image acquisition and processing we use a conventional upright stage microscope that is operated manually, the inexpensive Helicon Focus software, and the open source MeshLab software. Three objects were tested: an inclined flat glass slide with an imprinted 10 um calibration grid, a regular metal 100x100 per inch mesh, and a highly irregular surface of a material known as a porous electrode used in polyelectrolyte fuel cells. The accuracy of the reconstruction of the image features was found to be ~ 0.01 for the heights of ~ 100 um with the corresponding image acquisition time being ~ 10 min.