Detection Limits for SEM Image Segmentation

The dataset consists of six collections of SEM images, three trained U-net AI models, and CSV files with image quality metrics and trained AI model accuracy metrics. Each SEM image collection contains images augmented with Poisson noise and contrast.This work was performed with funding from the CHIPS Metrology Program, part of CHIPS for America, National Institute of Standards and Technology, U.S. Department of Commerce.

Data and Resources

Six sets of simulated SEM imageszip file with six subfolders containing SEM image collections
SEM images with varying noise and contrast values
Explore
- More information
- Go to resource
Six masks for the six simulated SEM image collectionszip file contains mask images and the initial intensity image with max contrast and min noise
image masks defining foreground and background labels
Explore
- More information
- Go to resource
Three trained AI models (U-net architecture)zip contains three folders with TensorFlow AI models
Three trained AI models on the first five of six simulated SEM image collections
Explore
- More information
- Go to resource
Tabular files with image and AI model metricszip file contains a folder with image quality metrics and a folder with AI model accuracy
Image quality metrics extracted from six simulated SEM images, as well as...
Explore
- More information
- Go to resource
relationships between AI model accuracy and...URL
graphs
Explore
- Preview
- Download
source code for detection limits
Python code
Explore
- More information
- Go to resource
READMETEXT
README.txt
Explore
- More information
- Go to resource

Field	Value
accessLevel	public
accrualPeriodicity	irregular
bureauCode	{006:55}
catalog_@context	https://project-open-data.cio.gov/v1.1/schema/data.json
catalog_conformsTo	https://project-open-data.cio.gov/v1.1/schema
catalog_describedBy	https://project-open-data.cio.gov/v1.1/schema/catalog.json
identifier	ark:/88434/mds2-3838
issued	2025-06-12
landingPage	https://data.nist.gov/od/id/mds2-3838
language	{en}
license	https://www.nist.gov/open/license
modified	2025-05-12 00:00:00
programCode	{006:045}
publisher	National Institute of Standards and Technology
resource-type	Dataset
source_datajson_identifier	true
source_hash	a20f7a04b55903c37b82a1250493c551b51d5eea43da260700fb69daa065f713
source_schema_version	1.1
theme	{"Information Technology:Computational science","Metrology:Dimensional metrology",Nanotechnology:Nanoelectronics,"Mathematics and Statistics:Image and signal processing","Mathematics and Statistics:Uncertainty quantification"}
Groups	AmeriGEOSS National Provider North America
Tags	AmeriGEO AmeriGEOSS CKAN GEO GEOSS National North America United States ai-model detection-limits dimensional-metrology scanning-electron-microscopy sem
isopen	False
license_id	other-license-specified
license_title	other-license-specified
maintainer	Peter Bajcsy
maintainer_email	peter.bajcsy@nist.gov
metadata_created	2025-09-23T15:07:04.024229
metadata_modified	2025-09-23T15:07:04.024237
notes	The dataset consists of six collections of SEM images, three trained U-net AI models, and CSV files with image quality metrics and trained AI model accuracy metrics. Each SEM image collection contains images augmented with Poisson noise and contrast.This work was performed with funding from the CHIPS Metrology Program, part of CHIPS for America, National Institute of Standards and Technology, U.S. Department of Commerce.
num_resources	7
num_tags	13
title	Detection Limits for SEM Image Segmentation