Multidimensional Statistical Patterns Observed on Water Drops Corona Discharge Pictures
Raveloson, Faniry1; Roussel, Jérémie1; Vandanjon, Laurent2*; de la Bardonnie,Hugues1
1 Laboratory LIMEC, 24 rue du Général Ferrié, 31500 Toulouse, France
2 Laboratory of Marine Biotechnology and Chemistry (LBCM), University Bretagne Sud (UBS), EMR CNRS 6076, IUEM, Campus Tohannic, 56000 Vannes, France
*Corresponding author: laurent.vandanjon@univ-ubs.fr
Keywords: Corona effect, machine learning algorithms, random forest, K-Nearest Neighbors (KNN), gradient Boost, decision tree, naive Bayes
Submitted: June 26, 2023
Reviewed: September 28, 2023
Accepted: October 10, 2023
Published: March 5, 2024
Abstract
Macroscopic corona images of droplets of different kinds of water were examined by collecting dozens of image parameters. A statistical comparison of these parameters using Data Science (machine learning) algorithms allowed us to differentiate water types with significant accuracy. The ability to differentiate water types using macroscopic corona imaging combined with machine learning algorithms presents topics for further studies.