Testing the Raman parameters of pollen spectra in automatic identification
- PDF / 764,144 Bytes
- 14 Pages / 547.087 x 737.008 pts Page_size
- 58 Downloads / 207 Views
(0123456789().,-volV) ( 01234567 89().,-volV)
ORIGINAL PAPER
Testing the Raman parameters of pollen spectra in automatic identification S. G. Pereira . A. Guedes . I. Abreu . H. Ribeiro
Received: 4 June 2020 / Accepted: 20 October 2020 Ó Springer Nature B.V. 2020
Abstract Pollen identification and quantification are used in many fields of application and research has been conducted to attain accurate automatic pollen recognition aiming to reduce the laborious work and subjectivity in human identification. The aim of our study was to evaluate the capacity of Raman parameters of pollen spectra, calculated for only 7 common band intervals in a limited spectral range, to be used as future technique in pollen automatic identification. There were analyzed 15 different pollen species considered to induce allergic reactions. Raman spectra were acquired at an excitation wavelength of 785 nm in a spectral region from 1000 to 1800 cm-1, preprocessed and deconvoluted to determine the Raman parameters: wavenumber, full width at half maximum
of the band and integrated intensity. Seven common band intervals of all Raman spectra, in the fingerprint areas 1000–1010, 1300–1460 and 1500–1700 cm-1, were chosen for the classification of the pollen species using SVM (support vector machine). Our results showed that the classification accuracy of all pollen species was 100% in the training step, while in the testing step 14 out of the 15 pollen species were correctly assigned (93.3%), including the discrimination between 5 Poaceae species and between Betula pendula and Corylus avellana. It was also observed that all Raman parameters are important in the classification as well as all wavenumber areas considered. So, our study indicates that the Raman parameters of pollen spectra can be a promising methodology for automatic pollen recognition.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10453-020-09669-1) contains supplementary material, which is available to authorized users.
Keywords Pollen classification Raman spectra Spectroscopy Support vector machine
S. G. Pereira A. Guedes H. Ribeiro (&) Department of Geosciences, Environment and Spatial Plannings, Faculty of Sciences, University of Porto, Rua do Campo Alegre, 687, 4169-007 Porto, Portugal e-mail: [email protected]
1 Introduction
A. Guedes I. Abreu H. Ribeiro Earth Sciences Institute (ICT), Pole of the Faculty of Sciences, University of Porto, Porto, Portugal I. Abreu Department of Biology of the Faculty of Sciences, University of Porto, Porto, Portugal
Pollen analysis has been used in many fields of application such as environmental monitoring (Ribeiro et al. 2015), agriculture (Cunha et al. 2016), paleobotany (Seddon et al. 2019; Schopf et al. 2016), forensic science (Orijemie and Israel 2019; Pereira et al. 2020) and medicine (Lo et al. 2019; Medek et al.
123
Aerobiologia
2019). Pollen is one of the most common triggers of season allergic reactions, in some individuals when inhaled causes sympto
Data Loading...
