Zernike moment based classification of cosmic ray candidate hits from cmos sensors

@article{58484,
   author = {Bar, O. and Bibrzycki, Ł. and Niedźwiecki, M. and Piekarczyk, M. and Rzecki, K. and Sośnicki, T. and Stuglik, S. and Frontczak, M. and Homola, P. and AlvarezandCastillo, D.E. and Andersen, T. and Tursunov, Arman},
   article_location = {Switzerland},
   article_number = {22},
   doi = {http://dx.doi.org/10.3390/s21227718},
   keywords = {CMOS sensors; feature-based classification; Zernike moments; machine learning; computer vision},
   language = {eng},
   issn = {1424-3210},
   journal = {Sensors},
   title = {Zernike moment based classification of cosmic ray candidate hits from cmos sensors},
   url = {https://www.mdpi.com/1424-8220/21/22/7718},
   volume = {21},
   year = {2021}
}

TY  - JOUR
ID  - 58484
AU  - Bar, O. - Bibrzycki, Ł. - Niedźwiecki, M. - Piekarczyk, M. - Rzecki, K. - Sośnicki, T. - Stuglik, S. - Frontczak, M. - Homola, P. - Alvarez-Castillo, D.E. - Andersen, T. - Tursunov, Arman
PY  - 2021
TI  - Zernike moment based classification of cosmic ray candidate hits from cmos sensors
JF  - Sensors
VL  - 21
IS  - 22
SP  - "7718-1"-"7718-18"
EP  - "7718-1"-"7718-18"
SN  - 14243210
KW  - CMOS sensors
KW  - feature-based classification
KW  - Zernike moments
KW  - machine learning
KW  - computer vision
UR  - https://www.mdpi.com/1424-8220/21/22/7718
N2  - Reliable tools for artefact rejection and signal classification are a must for cosmic ray detection experiments based on CMOS technology. In this paper, we analyse the fitness of several feature-based statistical classifiers for the classification of particle candidate hits in four categories: spots, tracks, worms and artefacts. We use Zernike moments of the image function as feature carriers and propose a preprocessing and denoising scheme to make the feature extraction more efficient. As opposed to convolution neural network classifiers, the feature-based classifiers allow for establishing a connection between features and geometrical properties of candidate hits. Apart from basic classifiers we also consider their ensemble extensions and find these extensions generally better performing than basic versions, with an average recognition accuracy of 88%.
ER  - 

BAR, O., Ł. BIBRZYCKI, M. NIED$\backslash$'ZWIECKI, M. PIEKARCZYK, K. RZECKI, T. SO$\backslash$'SNICKI, S. STUGLIK, M. FRONTCZAK, P. HOMOLA, D.E. ALVAREZ-CASTILLO, T. ANDERSEN and Arman TURSUNOV. Zernike moment based classification of cosmic ray candidate hits from cmos sensors. \textit{Sensors}. Switzerland, 2021, vol.~21, No~22, p.~''7718-1''-''7718-18'', 18 pp. ISSN~1424-3210. Available from: https://dx.doi.org/10.3390/s21227718.