View The Document

Accession Number:

AD1209124

Title:

Optical Imaging of Ultrafast Photo-Induced Phase Transitions of Nanostructures Using Neural Lenses

Author Organization(s):

• ROCHESTER UNIV NY

Report Date:

2021-08-09

Abstract:

Under this Short-Term Innovative Research (STIR) Project, we have made a range of progresses on our project on Machine Learning for dynamic surface structural reconstruction. Our main goal was to demonstrate machine learning's ability to recover images from optical degradation. We used Neural Networks (NNs) to apply image super-resolution to the images obtained using ultrafast time-resolved scattered-light microscopy techniques.

Document Type:

Conference:

Journal:

Pages:

11

File Size:

0.65MB

Contracts:

Grants:

Descriptors:

• machine learning
• optics
• femtosecond lasers
• light sources
• optical coatings
• laser beams
• learning
• microscopy
• phase transformations
• frequency
• lasers
• optical modulators
• femtosecond time
• films
• neural networks
• solar spectrum
• students

Identifiers:

• scattered light
• speckle noise
• gan(Generative Adversarial Networks)
• sttr reports
• sttr(small business technology transfer)

SubjectCategory:

• Mathematical and Computer Sciences

Communities of Interest:

• Autonomy

Distribution Statement:

Approved For Public Release

View The Document