Periodic feature characterization in nanostructured surfaces and emulsions

TL;DR

This paper presents a Python-based image analysis method for rapid, accurate measurement of surface features in nanostructured materials and emulsions, aiding the understanding of structure-function relationships.

Contribution

It introduces a new computational approach validated on synthetic and experimental images for characterizing complex surface features and morphology.

Findings

Manual measurements vary significantly compared to Python analysis.

The method can distinguish multiple feature populations within an image.

It accurately computes skewness and kurtosis of surface roughness.

Abstract

Understanding structure-function relationships is essential to advance the manufacturing of next-gen materials with desired properties and functionalities. Precise and rapid measurement of features like wrinkle size, droplet diameter, and surface roughness is essential to establishing such structure-function relationships. To this end, this work developed feature size and surface morphology characterizations through image analysis in Python and validated them with both synthetic and experimental images. Manual measurements of bio-based surfaces resulted in between 3.3% (N=50, visually simple) and 51.2% error (N=100, visually complex) compared to Python analysis results. This analysis was also used to accurately distinguish multiple feature size populations in a given image (which were missed entirely in manual measurements), and to determine the skewness and kurtosis of biological surfaces in a surface roughness map. This work contributes to a larger goal of developing a robust and computationally cheap platform to analyze complex materials to accelerate structure-function discovery.