Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Trainable segmentation for transmission electron microscope images of inorganic nanoparticles

Bell, Cameron G., Treder, Kevin P., Kim, Judy S., Schuster, Manfred E., Kirkland, Angus I. and Slater, Thomas J. A. ORCID: 2022. Trainable segmentation for transmission electron microscope images of inorganic nanoparticles. Journal of Microscopy 288 (3) , pp. 169-184. 10.1111/jmi.13110

[thumbnail of Journal of Microscopy - 2022 - Bell - Trainable segmentation for transmission electron microscope images of inorganic.pdf]
PDF - Published Version
Available under License Creative Commons Attribution.

Download (3MB) | Preview


We present a trainable segmentation method implemented within the python package ParticleSpy. The method takes user labelled pixels, which are used to train a classifier and segment images of inorganic nanoparticles from transmission electron microscope images. This implementation is based on the trainable Waikato Environment for Knowledge Analysis (WEKA) segmentation, but is written in python, allowing a large degree of flexibility and meaning it can be easily expanded using other python packages. We find that trainable segmentation offers better accuracy than global or local thresholding methods and requires as few as 100 user-labelled pixels to produce an accurate segmentation. Trainable segmentation presents a balance of accuracy and training time between global/local thresholding and neural networks, when used on transmission electron microscope images of nanoparticles. We also quantitatively investigate the effectiveness of the components of trainable segmentation, its filter kernels and classifiers, in order to demonstrate the use cases for the different filter kernels in ParticleSpy and the most accurate classifiers for different data types. A set of filter kernels is identified that are effective in distinguishing particles from background but that retain dissimilar features. In terms of classifiers, we find that different classifiers perform optimally for different image contrast; specifically, a random forest classifier performs best for high-contrast ADF images, but that QDA and Gaussian Naïve Bayes classifiers perform better for low-contrast TEM images.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Chemistry
Cardiff Catalysis Institute (CCI)
Additional Information: This is an open access article under the terms of the Creative Commons Attribution License
Publisher: Wiley
ISSN: 0022-2720
Funders: EPSRC
Date of First Compliant Deposit: 1 July 2022
Date of Acceptance: 26 April 2022
Last Modified: 24 May 2023 18:43

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics