HomoSeg: Deep Learning-Based Instance Segmentation with Object Tracking for Homopolymer Crystallization Kinetics

Guanjin Wang; Paul Roberts; Anthony Kotula; Kalman Migler; Peter Olmsted; Thao Nguyen; David Elbert

doi:10.34863/GMGW-0M79

HomoSeg: Deep Learning-Based Instance Segmentation with Object Tracking for Homopolymer Crystallization Kinetics

DOI 10.34863/GMGW-0M79 Published 2025 Type Poster Version 1.0

Open DOI View poster PDF

Publisher: PARADIM · Resource: MaRDA2025 Annual Meeting Poster Presentation

Abstract

MaRDA2025 Annual Meeting Best-Poster winner for Post-Doctoral Fellow category reporting streaming segmentation of in situ microscopy of polymer crystallization experiments using few-shot learning, data-driven model by a combination of deep learning-based segmentation and object tracking. The particle tracker is based on the DeepSORT algorithm to track the change of shaded masks of each instance across different frames.

Authors

Wang, Guanjin

Johns Hopkins University

Roberts, Paul

National Institute of Science and Technology · ORCID iD icon

https://orcid.org/0009-0007-5905-6198

Kotula, Anthony

National Institute of Science and Technology · ORCID iD icon

https://orcid.org/0000-0002-0830-2869

Migler, Kalman

National Institute of Science and Technology · ORCID iD icon

https://orcid.org/0000-0001-6538-3733

Olmsted, Peter

Georgetown University · ORCID iD icon

https://orcid.org/0000-0001-6538-9385

Nguyen, Thao

Johns Hopkins University · ORCID iD icon

https://orcid.org/0000-0002-0312-1583

Elbert, David

Hopkins Extreme Materials Institute · Johns Hopkins University · ORCID iD icon

https://orcid.org/0000-0002-2292-180X

Recommended citation

Wang, Guanjin, Roberts, Paul, Kotula, Anthony, Migler, Kalman, Olmsted, Peter, Nguyen, Thao, & Elbert, David (2025). HomoSeg: Deep Learning-Based Instance Segmentation with Object Tracking for Homopolymer Crystallization Kinetics (MaRDA2025 Annual Meeting Poster Presentation) [Poster]. PARADIM. https://doi.org/10.34863/GMGW-0M79

Record details

DOI

https://doi.org/10.34863/GMGW-0M79

Poster PDF

https://data.paradim.org/doi/gmgw-0m79/gmgw-0m79.pdf

Publisher

PARADIM

Publication year

2025

Resource type

Poster — MaRDA2025 Annual Meeting Poster Presentation

Subject

FOS: Materials engineering (Fields of Science and Technology (FOS))

Language

Version

1.0

License

Creative Commons Attribution Non Commercial No Derivatives 4.0 International
CC BY-NC-ND 4.0 badge

Funding

Funder

National Science Foundation

Award

2119040

Award title

DMREF/Collaborative Research: Integrated Material Design and Processing--Application to Recycled Plastics

Machine-readable metadata

Show DataCite XML

<?xml version="1.0" encoding="UTF-8"?>
<resource
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
    <identifier identifierType="DOI">10.34863/GMGW-0M79</identifier>
    <creators>
        <creator>
            <creatorName nameType="Personal">Wang, Guanjin</creatorName>
            <givenName>Guanjin</givenName>
            <familyName>Wang</familyName>
            <affiliation affiliationIdentifier="https://ror.org/00za53h95" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">Johns Hopkins University</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Roberts, Paul</creatorName>
            <givenName>Paul</givenName>
            <familyName>Roberts</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0009-0007-5905-6198</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/01150px97" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">National Institute of Science and Technology</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Kotula, Anthony</creatorName>
            <givenName>Anthony</givenName>
            <familyName>Kotula</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0000-0002-0830-2869</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/01150px97" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">National Institute of Science and Technology</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Migler, Kalman</creatorName>
            <givenName>Kalman</givenName>
            <familyName>Migler</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0000-0001-6538-3733</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/01150px97" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">National Institute of Science and Technology</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Olmsted, Peter</creatorName>
            <givenName>Peter</givenName>
            <familyName>Olmsted</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0000-0001-6538-9385</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/05vzafd60" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">Georgetown University</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Nguyen, Thao</creatorName>
            <givenName>Thao</givenName>
            <familyName>Nguyen</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0000-0002-0312-1583</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/00za53h95" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">Johns Hopkins University</affiliation>
        </creator>
        <creator>
            <creatorName nameType="Personal">Elbert, David</creatorName>
            <givenName>David</givenName>
            <familyName>Elbert</familyName>
            <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="https://orcid.org">https://orcid.org/0000-0002-2292-180X</nameIdentifier>
            <affiliation affiliationIdentifier="https://ror.org/02ed2th17" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">Hopkins Extreme Materials Institute</affiliation>
            <affiliation affiliationIdentifier="https://ror.org/00za53h95" affiliationIdentifierScheme="ROR" schemeURI="https://ror.org">Johns Hopkins University</affiliation>
        </creator>
    </creators>
    <titles>
        <title xml:lang="en">HomoSeg: Deep Learning-Based Instance Segmentation with Object Tracking for Homopolymer Crystallization Kinetics</title>
    </titles>
    <publisher publisherIdentifier="https://ror.org/04pw1zg89" publisherIdentifierScheme="ROR" schemeURI="https://ror.org">PARADIM</publisher>
    <publicationYear>2025</publicationYear>
    <resourceType resourceTypeGeneral="Poster">MaRDA2025 Annual Meeting Poster Presentation</resourceType>
    <subjects>
        <subject subjectScheme="Fields of Science and Technology (FOS)" schemeURI="https://web-archive.oecd.org/2012-06-15/138575-38235147.pdf" classificationCode="2.5">FOS: Materials engineering</subject>
    </subjects>
    <language>en</language>
    <sizes/>
    <formats/>
    <version>1.0</version>
    <rightsList>
        <rights rightsURI="https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode">Creative Commons Attribution Non Commercial No Derivatives 4.0 International</rights>
    </rightsList>
    <fundingReferences>
        <fundingReference>
            <funderName>National Science Foundation</funderName>
            <funderIdentifier funderIdentifierType="Crossref Funder ID">https://doi.org/10.13039/100000001</funderIdentifier>
            <awardNumber awardURI="https://www.nsf.gov/awardsearch/show-award?AWD_ID=2119040">2119040</awardNumber>
            <awardTitle>DMREF/Collaborative Research: Integrated Material Design and Processing--Application to Recycled Plastics</awardTitle>
        </fundingReference>
    </fundingReferences>
</resource>