Where is the Machine Looking? : Locating Discriminative Light-Scattering Features by Class-Activation Mapping

Show full item record



Permalink

http://hdl.handle.net/10138/315734

Citation

Piedra , P , Gobert , C , Kalume , A , Pan , Y-L , Kocifaj , M , Muinonen , K , Penttilä , A , Zubko , E & Videen , G 2020 , ' Where is the Machine Looking? Locating Discriminative Light-Scattering Features by Class-Activation Mapping ' , Journal of Quantitative Spectroscopy & Radiative Transfer , vol. 247 , 106936 . https://doi.org/10.1016/j.jqsrt.2020.106936

Title: Where is the Machine Looking? : Locating Discriminative Light-Scattering Features by Class-Activation Mapping
Author: Piedra, Patricio; Gobert, Christian; Kalume, Aimable; Pan, Yong-Le; Kocifaj, Miroslav; Muinonen, Karri; Penttilä, Antti; Zubko, Evgenij; Videen, Gorden
Contributor: University of Helsinki, Department of Physics
University of Helsinki, Department of Physics
Date: 2020-05
Language: eng
Number of pages: 8
Belongs to series: Journal of Quantitative Spectroscopy & Radiative Transfer
ISSN: 0022-4073
URI: http://hdl.handle.net/10138/315734
Abstract: We explore a technique called class-activation mapping (CAM) to investigate how a Machine Learning (ML) architecture learns to classify particles based on their light-scattering signals. We release our code, and also find that different regions of the light-scattering signals play different roles in ML classification. These regions depend on the type of particles being classified and on the nature of the data obtained and trained. For instance, the Mueller-matrix elements S11*,S12* and S21*had the greatest classification activation in the diffraction region. Linear polarization elements S12* and S21* were most accurate in the backscattering region for clusters of spheres and spores, and was most accurate in the diffraction region for other particle classes. The CAM technique was able to highlight specific light-scattering angles that maximize the potential for discrimination of extremely similar particle classes. Such information is useful for designing detector systems to classify particles where limited space or resources are available, including flow cytometry and satellite remote sensing.
Subject: Light scattering
Deep learning
114 Physical sciences
Rights:


Files in this item

Total number of downloads: Loading...

Files Size Format View
1_s2.0_S0022407319307149_main.pdf 2.590Mb PDF View/Open
1_s2.0_S0022407319307149_main.pdf 1.139Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record