Related Experiment Video
Updated: Jan 6, 2026
Studying the Effects of Temperature on the Nucleation and Growth of Nanoparticles by Liquid-Cell Transmission Electron Microscopy
Published on: February 17, 2021
A new method for measuring nanoparticle diameter from a set of SEM images using a remarkable point
Loïc Crouzier1, Alexandra Delvallée2, Sébastien Ducourtieux2
1Laboratoire National de métrologie et d'Essais - Nanometrology, 29 avenue Hennequin, 78197 Trappes Cedex, France; Institut Pprime Département Physique et Mécanique des Matériaux - 11 Bd Marie et Pierre Curie, 86962 Futuroscope Chasseneuil, France.
A new method using a "remarkable point" accurately extracts nanoparticle boundaries from Scanning Electron Microscopy (SEM) images. This technique significantly reduces measurement uncertainty compared to conventional segmentation methods.
Area of Science:
- Materials Science
- Nanotechnology
- Metrology
Background:
- Scanning Electron Microscopy (SEM) is a key technique for nanoparticle (NP) dimensional analysis.
- Accurate NP segmentation from SEM images is crucial for reliable measurements.
- Conventional segmentation methods often lack precision and have associated uncertainties.
Purpose of the Study:
- To introduce a novel method for extracting NP boundaries from SEM images using a remarkable point.
- To evaluate the robustness and accuracy of this new method against conventional techniques.
- To reduce measurement uncertainty in NP dimensional analysis via SEM.
Main Methods:
- Theoretical study using Monte Carlo simulations on silica NPs.
- Comparison with conventional segmentation techniques like Active Contour and Full Width at Half-Maximum (FWHM) binarization.
- Application of the remarkable point method to SEM images of silica reference nanoparticles.
Main Results:
- Conventional methods showed systematic discrepancies and size-dependent errors in NP measurements.
- The remarkable point method demonstrated excellent agreement between measured and simulated NP diameters.
- Segmentation accuracy was validated using silica reference nanoparticles, matching calibration certificate data.
Conclusions:
- The remarkable point method accurately delimits NP contours in SEM images.
- This novel approach significantly reduces measurement uncertainty, from 4.3 nm to 2.6 nm (k=2).
- The remarkable point method offers a more reliable approach for NP dimensional metrology using SEM.
