Classification of Fungi Species by Hyperspectral Images using Machine Learning

• Adriel L. V. Mori UFG
• Arlindo R. Galvão Filho UFG

DOI: https://doi.org/10.5753/sbcas_estendido.2024.1515

Abstract

This work proposes the classification of four species of fungi using Hyperspectral imaging (HSI) and Machine Learning (ML). The HSI technique, which acquires spectral data quickly and non-destructively, is used to improve the identification of fungal species and automate manual processes in microbiological products. The study focuses on the differentiation of fungal species using HSI and supervised machine learning, achieving an accuracy of 97.12% with the Multilayer Perceptron (MLP) neural network classifier. The results highlight the potential of using ML and HSI in the differentiation of fungal species in clinical and microbiological environments.

References

Arrigoni, S., Turra, G., and Signoroni, A. (2017). Hyperspectral image analysis for rapid and accurate discrimination of bacterial infections: A benchmark study. Computers in Biology and Medicine, 88:60–71.

Gewali, U. B., Monteiro, S. T., and Saber, E. (2018). Machine learning based hyperspectral image analysis: a survey. arXiv preprint arXiv:1802.08701.

Gowen, A. A., Feng, Y., Gaston, E., and Valdramidis, V. (2015). Recent applications of hyperspectral imaging in microbiology. Talanta, 137:43–54.

Kingma, D. P. and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.

Kokka, A., Pulli, T., Honkavaara, E., Markelin, L., Kärhä, P., and Ikonen, E. (2019). Flat-field calibration method for hyperspectral frame cameras. Metrologia, 56(5):055001.

Lu, Y., Wang, W., Huang, M., Ni, X., Chu, X., and Li, C. (2020). Evaluation and classification of five cereal fungi on culture medium using visible/near-infrared (vis/nir) hyperspectral imaging. Infrared Physics & Technology, 105:103206.

Savitzky, A. and Golay, M. J. (1964). Smoothing and differentiation of data by simplified least squares procedures. Analytical chemistry, 36(8):1627–1639.

Sibirny, A. A., editor (2017). Biotechnology of yeasts and filamentous fungi. Springer.

Stahl, P. D. and Klug, M. J. (1996). Characterization and differentiation of filamentous fungi based on fatty acid composition. Applied and Environmental Microbiology, 62(11):4136–4146.

Williams, P. J., Geladi, P., Britz, T. J., and Manley, M. (2012). Growth characteristics of three fusarium species evaluated by near-infrared hyperspectral imaging and multivariate image analysis. Applied microbiology and biotechnology, 96:803–813.

Yao, H., Hruska, Z., DiCrispino, K., Brabham, K., Lewis, D., Beach, J., Brown, R. L., and Cleveland, T. E. (2005). Differentiation of fungi using hyperspectral imagery for food inspection. In 2005 ASAE Annual Meeting, page 1. American Society of Agricultural and Biological Engineers.

Yao, H., Hruska, Z., Kincaid, R., Brown, R. L., and Cleveland, T. E. (2008). Differentiation of toxigenic fungi using hyperspectral imagery. Sensing and Instrumentation for Food Quality and Safety, 2:215–224.