Return to search results
A Data-Driven Approach to Complex Voxel Predictions in Grayscale Digital Light Processing Additive Manufacturing Using U-nets and Generative Adversarial Networks
Digital light processing (DLP) vat photopolymerization (VP) additive manufacturing (AM) uses patterned UV light to selectively cure a liquid photopolymer into a solid layer. Subsequent layers are printed on to preceding layers to eventually form a desired 3 dimensional (3D) part. This data set characterizes the 3D geometry of a single layer of voxels (volume pixels) printed with photomasks assigned random intensity levels at every pixel. The masks are computer generated, then printed onto a glass cover slide. Geometry of the printed voxels is characterized by laser scanning confocal microscopy. The data were originally curated to train image-to-image U-net machine learning models to predict voxel scale geometry given arbitrary photomasks, as described in the publication "A Data-Driven Approach to Complex Voxel Predictions in Grayscale Digital Light Processing Additive Manufacturing Using U-nets and Generative Adversarial Networks". Data are provided in a raw (native microscope format and photomask image) and processed into aligned mask-print training pairs. A total of 1500 8 pixel × 8 pixel (i.e. 96 000 pixel interactions) training pairs are provided. Jupyter notebooks for various steps in process are also provided.
Complete Metadata
| @type | dcat:Dataset |
|---|---|
| accessLevel | public |
| accrualPeriodicity | irregular |
| bureauCode |
[
"006:55"
]
|
| contactPoint |
{
"fn": "Jason Killgore",
"hasEmail": "mailto:jason.killgore@nist.gov"
}
|
| description | Digital light processing (DLP) vat photopolymerization (VP) additive manufacturing (AM) uses patterned UV light to selectively cure a liquid photopolymer into a solid layer. Subsequent layers are printed on to preceding layers to eventually form a desired 3 dimensional (3D) part. This data set characterizes the 3D geometry of a single layer of voxels (volume pixels) printed with photomasks assigned random intensity levels at every pixel. The masks are computer generated, then printed onto a glass cover slide. Geometry of the printed voxels is characterized by laser scanning confocal microscopy. The data were originally curated to train image-to-image U-net machine learning models to predict voxel scale geometry given arbitrary photomasks, as described in the publication "A Data-Driven Approach to Complex Voxel Predictions in Grayscale Digital Light Processing Additive Manufacturing Using U-nets and Generative Adversarial Networks". Data are provided in a raw (native microscope format and photomask image) and processed into aligned mask-print training pairs. A total of 1500 8 pixel × 8 pixel (i.e. 96 000 pixel interactions) training pairs are provided. Jupyter notebooks for various steps in process are also provided. |
| distribution |
[
{
"title": "2950_README",
"mediaType": "text/plain",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/2950_README.txt"
},
{
"title": "Jupyter_notebooks",
"mediaType": "application/zip",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/Jupyter_notebooks.zip"
},
{
"title": "modified_pix2pix",
"mediaType": "application/zip",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/modified_pix2pix.zip"
},
{
"title": "photomasks",
"mediaType": "application/zip",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/photomasks.zip"
},
{
"title": "raw_print_data",
"mediaType": "application/zip",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/raw_print_data.zip"
},
{
"title": "training_pairs",
"mediaType": "application/zip",
"downloadURL": "https://data.nist.gov/od/ds/mds2-2950/training_pairs.zip"
}
]
|
| identifier | ark:/88434/mds2-2950 |
| issued | 2023-07-20 |
| keyword |
[
"3D Printing",
"Additive Manufacturing",
"Generative Adversarial Network",
"Machine Learning",
"Photopolymer"
]
|
| landingPage | https://data.nist.gov/od/id/mds2-2950 |
| language |
[
"en"
]
|
| license | https://www.nist.gov/open/license |
| modified | 2023-03-07 00:00:00 |
| programCode |
[
"006:045"
]
|
| publisher |
{
"name": "National Institute of Standards and Technology",
"@type": "org:Organization"
}
|
| references |
[
"https://doi.org/10.1002/smll.202301987"
]
|
| theme |
[
"Manufacturing:Additive manufacturing",
"Materials:Polymers",
"Mathematics and Statistics:Statistical analysis"
]
|
| title | A Data-Driven Approach to Complex Voxel Predictions in Grayscale Digital Light Processing Additive Manufacturing Using U-nets and Generative Adversarial Networks |
