A4 Conference proceedings

Automated Segmentation of Nanoparticles in BF TEM Images by U-Net Binarization and Branch and Bound

Open Access publication

Publication Details
Authors: Zafari Sahar, Eerola Tuomas, Ferreira Paulo, Kälviäinen Heikki, Bovik Alan
Publisher: Springer Verlag (Germany): Series
Publication year: 2019
Language: English
Related Journal or Series Information: Lecture Notes in Computer Science
Title of parent publication: Computer Analysis of Images and Patterns, Springer Lecture Notes in Computer Science
Journal acronym: LNCS
Volume number: 11678
Start page: 113
End page: 125
Number of pages: 13
ISBN: 978-3-030-29887-6
eISBN: 978-3-030-29888-3
ISSN: 0302-9743
eISSN: 1611-3349
JUFO-Level of this publication: 1
Open Access: Open Access publication
Location of the parallel saved publication: http://urn.fi/URN:NBN:fi-fe2019102534808

Transmission electron microscopy (TEM) provides information about inorganic nanoparticles that no other method is able to deliver. Yet, a major task when studying Inorganic nanoparticles using TEM is the automated analysis of the images, i.e. segmentation of individual nanoparticles. The current state-of-the-art methods generally rely on binarization routines that require parameter-ization, and on methods to segment the overlapping nanoparticles (NPs) usinghighly idealized nanoparticle shape models. It is unclear, however, that there is any way to determine the best set of parameters providing an optimal segmentation, given the great diversity of NPs characteristics, such as shape and size, that may be encountered. Towards remedying these barriers, this paper introduces a method for segmentation of NPs in Bright Field (BF) TEM images. The proposed method involves three main steps: binarization, contour evidence extraction, and contour estimation. For the binarization, a model based on the U-Net architec-ture is trained to convert an input image into its binarized version. The contour evidence extraction starts by recovering contour segments from a binarized im-age using concave contour points detection. The contour segments which belong to the same nanoparticle are grouped in the segment grouping step. The group-ing is formulated as a combinatorial optimization problem and solved using thewell-known branch and bound algorithm. Finally, the full contours of the NPsare estimated by an ellipse. The experiments on a real-world dataset consisting of 150 BF TEM images containing approximately 2,700 NPs show that the proposed method outperforms five current state-of-art approaches in the overlapping NPs segmentation.

Last updated on 2020-20-03 at 10:03