Preprints
https://doi.org/10.5194/amt-2022-72
https://doi.org/10.5194/amt-2022-72
 
03 May 2022
03 May 2022
Status: this preprint is currently under review for the journal AMT.

Ice crystals images from Optical Array Probes: classification with Convolutional Neural Networks

Louis Jaffeux, Alfons Schwarzenböck, Pierre Coutris, and Christophe Duroure Louis Jaffeux et al.
  • Laboratoire de Météorologie Physique (UMR6016) / UCA / CNRS, Aubière, France

Abstract. Although airborne optical array probes (OAP) have existed for decades, our ability to maximize extraction of meaningful morphological information out of the images produced by these probes has been limited by the lack of automatic, unbiased and reliable classification tools. The present study describes a methodology for automatic ice crystal recognition using innovative machine learning. Convolutional Neural Network (CNN) have recently been perfected for computer vision and have been chosen as the method to achieve the best results together with the use of finely tuned dropout layers. For the purposes of this study, CNN has been adapted for the Precipitation Imaging Probe (PIP) and the 2DS-Stereo Probe (2DS), two commonly used probes that differ in pixel resolution and measurable maximum size range for hydrometeors. Six morphological crystal classes have been defined for the PIP and eight crystal classes and an artifact class, for the 2DS. The PIP and 2DS classifications have five common classes. In total more than 8000 images from both instruments have been manually labelled, thus allowing for the initial training. For each probe the classification design tries to account for the three primary ice crystal growth processes: vapor deposition, riming and aggregation. We included classes such as fragile aggregates and rimed aggregates with high intra-class shape variability and commonly found in convective clouds. The trained network is finally tested through human random inspections on actual data to show its real performance in comparison to what humans can achieve.

Louis Jaffeux et al.

Status: open (until 23 Jun 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Louis Jaffeux et al.

Louis Jaffeux et al.

Viewed

Total article views: 172 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
135 36 1 172 2 1
  • HTML: 135
  • PDF: 36
  • XML: 1
  • Total: 172
  • BibTeX: 2
  • EndNote: 1
Views and downloads (calculated since 03 May 2022)
Cumulative views and downloads (calculated since 03 May 2022)

Viewed (geographical distribution)

Total article views: 180 (including HTML, PDF, and XML) Thereof 180 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 25 May 2022
Download
Short summary
Optical Array Probes are instruments used aboard research aircrafts to capture 2D images of ice or water particles in clouds. This study presents a new tool using innovative machine learning, called convolutionnal neural networks, designed to identify the shape of imaged ice particles for two of these imagers, namely 2DS and PIP. Such a tool will be a very strong asset for understanding cloud microphysics. Beyond traditional evaluation metrics, human inspections were performed on unknown data.