A multistage, semi-automated procedure for analyzing the morphology of nanoparticles

Abstract

This article presents a multistage, semi-automated procedure that can expedite the morphology analysis of nanoparticles. Material scientists have long conjectured that the morphology of nanoparticles has a profound impact on the properties of the hosting material, but a bottleneck is the lack of a reliable and automated morphology analysis of the particles based on their image measurements. This article attempts to fill in this critical void. One particular challenge in nanomorphology analysis is how to analyze the overlapped nanoparticles, a problem not well addressed by the existing methods but effectively tackled by the method proposed in this article. This method entails multiple stages of operations, executed sequentially, and is considered semi-automated due to the inclusion of a semi-supervised clustering step. The proposed method is applied to several images of nanoparticles, producing the needed statistical characterization of their morphology.

Keywords:

• Nano imaging
• morphology analysis
• shape analysis
• machine learning
• nanoparticle overlapping

Acknowledgements

The authors would like to acknowledge the generous support from their sponsors. Ding and Park were partially supported by NSF grants CMMI-0348150 and CMMI-1000088; Huang was partially supported by NSF grants DMS-0606580, and DMS-0907170; Ding, Park, Mallick, and Liang were partially supported by Texas Norman Hackerman Advanced Research Program grant 010366-0024-2007; Huang, Kundu, and Mallick were partially supported by King Abdullah University of Science and Technology award KUS-CI-016-04.