Atmospheric Measurement Techniques
Atmospheric Measurement Techniques
AMT
Articles | Volume 18, issue 21
Articles | Volume 18, issue 21
https://doi.org/10.5194/amt-18-6291-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/amt-18-6291-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 18, issue 21
Research article
 | 
06 Nov 2025
Research article |  | 06 Nov 2025

In-line holographic droplet imaging: accelerated classification with convolutional neural networks and quantitative experimental validation

Birte Thiede, Oliver Schlenczek, Katja Stieger, Alexander Ecker, Eberhard Bodenschatz, and Gholamhossein Bagheri

Viewed

Total article views: 5,458 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
3,552 1,751 155 5,458 126 224
  • HTML: 3,552
  • PDF: 1,751
  • XML: 155
  • Total: 5,458
  • BibTeX: 126
  • EndNote: 224
Views and downloads (calculated since 03 Apr 2025)
Cumulative views and downloads (calculated since 03 Apr 2025)

Viewed (geographical distribution)

Total article views: 5,458 (including HTML, PDF, and XML) Thereof 5,362 with geography defined and 96 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Saved (final revised paper)

Latest update: 20 Jul 2026
Download
Short summary
Accurate measurement of cloud particles is crucial for cloud research. While holographic imaging enables detailed analysis of cloud droplet size, shape, and distribution, processing errors remain poorly quantified. To address this, we developed CloudTarget, a patterned photomask that can quantify the detection efficiency and uncertainties. Additionally, our AI-based classification enhances both accuracy and speed, achieving over 90 % precision while accelerating analysis 100-fold.
Read more
Share