Publikationstyp: | Konferenz: Paper |
Art der Begutachtung: | Peer review (Publikation) |
Titel: | Understanding the learning mechanism of convolutional neural networks applied to fluorescence spectra |
Autor/-in: | Venturini, Francesca Michelucci, Umberto Sperti, Michela Gucciardi, Arnaud Deriu, Marco A. |
et. al: | No |
DOI: | 10.1117/12.2647809 |
Tagungsband: | AI and Optical Data Sciences IV |
Herausgeber/-in des übergeordneten Werkes: | Jalali, Bahram Kitayama, Ken-ichi |
Band(Heft): | 124380 |
Angaben zur Konferenz: | SPIE Photonics West, San Francisco, USA, 28 January - 2 February 2023 |
Erscheinungsdatum: | Mär-2023 |
Verlag / Hrsg. Institution: | SPIE |
ISBN: | 9781510659810 9781510659827 |
Sprache: | Englisch |
Schlagwörter: | Fluorescence spectroscopy; Olive oil; Machine learning; Artificial neural network; Quality control; Explainability; Convolutional neural network |
Fachgebiet (DDC): | 006: Spezielle Computerverfahren 530: Physik |
Zusammenfassung: | The power of artificial neural networks to determine the quality and properties of olive oil was proven by several studies in the last years. Less clear is, however, how the neural network is able to extract useful information from the input data. This work investigates the learning mechanism of one-dimensional convolutional neural networks (1D-CNNs) trained to predict the physicochemical properties of olive oil from single fluorescence spectra. Such a 1D-CNN can successfully predict the parameters relevant to the quality assessment: acidity, peroxide value, and UV absorbance. To go beyond a simple quality assessment algorithm, it is important to identify which spectral features in the measured spectra are correlated with each chemical parameter and therefore with the quality of olive oil. To obtain this information, explainability techniques can be used by studying the latent feature space generated by the intermediate layers of the one-dimensional trained convolutional neural network. This work analyses in detail the common features that are used by the 1D-CNN to predict the two physicochemical parameters: acidity and K232. |
URI: | https://digitalcollection.zhaw.ch/handle/11475/27607 |
Volltext Version: | Publizierte Version |
Lizenz (gemäss Verlagsvertrag): | Lizenz gemäss Verlagsvertrag |
Departement: | School of Engineering |
Organisationseinheit: | Institut für Angewandte Mathematik und Physik (IAMP) |
Publiziert im Rahmen des ZHAW-Projekts: | Self-learning optical sensor |
Enthalten in den Sammlungen: | Publikationen School of Engineering |
Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.
Zur Langanzeige
Venturini, F., Michelucci, U., Sperti, M., Gucciardi, A., & Deriu, M. A. (2023). Understanding the learning mechanism of convolutional neural networks applied to fluorescence spectra [Conference paper]. In B. Jalali & K.-i. Kitayama (Eds.), AI and Optical Data Sciences IV (Vol. 124380). SPIE. https://doi.org/10.1117/12.2647809
Venturini, F. et al. (2023) ‘Understanding the learning mechanism of convolutional neural networks applied to fluorescence spectra’, in B. Jalali and K.-i. Kitayama (eds) AI and Optical Data Sciences IV. SPIE. Available at: https://doi.org/10.1117/12.2647809.
F. Venturini, U. Michelucci, M. Sperti, A. Gucciardi, and M. A. Deriu, “Understanding the learning mechanism of convolutional neural networks applied to fluorescence spectra,” in AI and Optical Data Sciences IV, Mar. 2023, vol. 124380. doi: 10.1117/12.2647809.
VENTURINI, Francesca, Umberto MICHELUCCI, Michela SPERTI, Arnaud GUCCIARDI und Marco A. DERIU, 2023. Understanding the learning mechanism of convolutional neural networks applied to fluorescence spectra. In: Bahram JALALI und Ken-ichi KITAYAMA (Hrsg.), AI and Optical Data Sciences IV. Conference paper. SPIE. März 2023. ISBN 9781510659810
Venturini, Francesca, Umberto Michelucci, Michela Sperti, Arnaud Gucciardi, and Marco A. Deriu. 2023. “Understanding the Learning Mechanism of Convolutional Neural Networks Applied to Fluorescence Spectra.” Conference paper. In AI and Optical Data Sciences IV, edited by Bahram Jalali and Ken-ichi Kitayama. Vol. 124380. SPIE. https://doi.org/10.1117/12.2647809.
Venturini, Francesca, et al. “Understanding the Learning Mechanism of Convolutional Neural Networks Applied to Fluorescence Spectra.” AI and Optical Data Sciences IV, edited by Bahram Jalali and Ken-ichi Kitayama, vol. 124380, SPIE, 2023, https://doi.org/10.1117/12.2647809.
Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.