Behind Density Lines: An Interface to Manually Segment Scanning Electron Microscopy Images

Simple item page
dc.contributor.authorNguyen, Anthonyen
dc.contributor.authorDiGiovanna, Lukeen
dc.contributor.authorSiegel, Johnen
dc.contributor.authorLin, Alexen
dc.date.accessioned2023-12-08T02:07:35Zen
dc.date.available2023-12-08T02:07:35Zen
dc.date.issued2023-11-30en
dc.descriptionBehindDensityLines.zip - Entire code repository for the project. BehindDensityLinesPresentation.pdf - Final project presentation in PDF. BehindDensityLinesPresentation.pptx - Final project presentation in PowerPoint. BehindDensityLinesReport.pdf - Final project report.en
dc.description.abstractSEM (Scanning Electron Microscopy) is a strong imaging technique used in many scientific domains, including materials science, biology, and nanotechnology. Researchers can use SEM to obtain high-resolution images of specimen surface morphology and topography, providing a precise glimpse of structures at the nanoscale. SEM photos reveal intricate surface details, allowing scientists to investigate the texture, shape, and size of particles, cells, or materials with incredible accuracy. Currently, manual segmentation of SEM images is an important stage in the analysis process for researchers. Manual segmentation entails painstakingly drawing and naming sections of interest within photographs, such as specific structures or particles. Researchers often trace object boundaries using sophisticated software tools built for picture processing and analysis. We built a gameified multiplayer online application allowing individual contributors to manually segment a SEM picture in real time due to the time and effort required. One important goal was to involve the next generation of scientists and researchers with a demonstration at the Virginia Tech Science Festival in November 2023. We designed a comparison score technique for a given segmentation to a reference segmentation for a specific SEM picture to provide participants with fast feedback. This enabled individuals and groups to measure their performance while also incorporating a gaming element. We now have a comprehensive understanding of how to create a full-stack project thanks to this initiative. We discovered how to leverage Amazon Web Services, such as EC2, to scale the infrastructure of our website from the backend. Through the use of Javascript frameworks and packages such as NextJS, Socket.io, and ThreeJS, we have created an intuitive user interface for group manual segmentation.en
dc.identifier.urihttps://hdl.handle.net/10919/117110en
dc.language.isoen_USen
dc.publisherVirginia Techen
dc.subjectSEMen
dc.subjectScanning Electron Microscopyen
dc.subjectSEM Segmentingen
dc.subjectImage Segmentationen
dc.subjectWeb Applicationen
dc.subjectWeb Developmenten
dc.subjectVirginia Tech Science Festivalen
dc.subjectPainter Canvas Appen
dc.titleBehind Density Lines: An Interface to Manually Segment Scanning Electron Microscopy Imagesen
dc.typeReporten
dc.typePresentationen

Files

Original bundle
Now showing 1 - 4 of 4
Name:
BehindDensityLines.zip
Size:
70 MB
Format:
Download
Thumbnail Image
Name:
BehindDensityLinesPresentation.pdf
Size:
3.02 MB
Format:
Adobe Portable Document Format
Download
Name:
BehindDensityLinesPresentation.pptx
Size:
21.12 MB
Format:
Microsoft Powerpoint XML
Download
Thumbnail Image
Name:
BehindDensityLinesReport.pdf
Size:
2.86 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
Name:
license.txt
Size:
1.5 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

CS4624: Multimedia, Hypertext, and Information Access