A bottom-up approach to select constrained spectral bands discriminating vine diseases

• Autorzy: Zhang Shurong , Goupil Alban , Vrabie Valeriu , Perrin Eric , Panon Marie-Laure

Identyfikatory

DOI

10.15439/2024F7286

• Konferencja: Federated Conference on Computer Science and Information Systems (19 ; 08-11.09.2024 ; Belgrade, Serbia)
• Języki publikacji: EN

Abstrakty

EN

The detection and control of diseases constitute a primary objective of French viticultural research. In this paper, we present a bottom-up hierarchical approach for selecting spectral bands suitable for class discrimination of spectra acquired by Infrared spectroscopy. Our method entails evaluating neighbouring bands using various similarity metrics, applying aggregation criteria, and ultimately identifying a limited number of the most relevant bands for the separation of classes. The bandwidths are limited within a range as is typically required for choosing existing optical filters or specifying coloured filter arrays. Our approach facilitates the discovery of distinctive spectral bands associated with a disease of interest, enabling the customization of multispectral cameras to meet specific requirements. It was applied to spectra collected on vine leaves spanning a three-year period with the goal to identify the most discriminant bands for the detection of grapevine yellows. The results show that a limited number of bands are sufficient to identify this class of interest through a classifier based on Linear Discriminant Analysis.

Słowa kluczowe

EN

band selection; hierarchical model; classification; grapevine flavescence dorée

PL

model hierarchiczny; klasyfikacja; złota żółtaczka; choroby winorośli

• Wydawca: Polskie Towarzystwo Informatyczne
• Czasopismo: Annals of Computer Science and Information Systems
• Rocznik: 2024
• Tom: Vol. 39
• Strony: 707--712
• Opis fizyczny: Bibliogr. 7 poz., tab., wykr., wz.

Twórcy

autor

Zhang Shurong

• Université de Reims Champagne-Ardenne, CReSTIC, Reims, France

autor

Goupil Alban

• Université de Reims Champagne-Ardenne, CReSTIC, Reims, France

autor

Vrabie Valeriu

• Université de Reims Champagne-Ardenne, CReSTIC, Reims, France

autor

Perrin Eric

• Université de Reims Champagne-Ardenne, CReSTIC, Reims, France

autor

Panon Marie-Laure

• Comité Champagne, 5 Rue Henri Martin, 51200 Épernay, France

Bibliografia

• 1. A. Martínez-Usó, F. Pla, J. M. Sotoca, P. García-Sevilla, Clustering-Based Hyperspectral Band Selection Using Information Measures, IEEE Trans. Geosci. Remote Sens. 45(12), pp. 4158-4171, Dec. 2007. http://dx.doi.org/10.1109/TGRS.2007.904951
• 2. H. Sun, L. Zhang, J. Ren, H. Huang, Novel hyperbolic clustering-based band hierarchy (HCBH) for effective unsupervised band selection of hyperspectral images, Pattern Recognition, 130 :108788, October 2022. http://dx.doi.org/10.1016/j.patcog.2022.108788
• 3. H. Peng, F. Long, C. Ding, Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy, IEEE Trans. Pattern Anal. Mach. Intell. 27(8), pp. 1226-1238, Aug. 2005. http://dx.doi.org/10.1109/TPAMI.2005.159
• 4. P. Granitto, C. Furlanello, F. Biasioli, F. Gasperi, Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial products, Chemometrics and Intelligent Laboratory Systems, 83(2), 2006, http://dx.doi.org/10.1016/j.chemolab.2006.01.007
• 5. D. Aha, R. Bankert, A comparative evaluation of sequential feature selection algorithms, Proc. Fifth International Workshop on Artificial Intelligence and Statistics, PMLR, 1995.
• 6. S. Zhang, E. Perrin, A. Goupil, V. Vrabie, M.-L. Panon, Nouveau modèle hiérarchique ascendant pour la sélection des bandes spectrales discriminant les maladies de la vigne, Colloque GRETSI, Grenoble, 2023.
• 7. U. V. Luxburg, A tutorial on spectral clustering, Statistics and Computing, 17(4):395–416, December 2007. http://dx.doi.org/10.48550/arXiv.0711.0189

Uwagi

Thematic Sessions: Short Papers