Journal of Spectral Imaging Abstract

Journal of Spectral Imaging,   Volume 7   Article ID a8   (2018)

Peer reviewed paper

Near infrared hyperspectral images and pattern recognition techniques used to identify etiological agents of cotton anthracnose and ramulosis

  • Priscila S.R. Aires
  • Francisco F. Gambarra-Neto
  • Wirton M. Coutinho
  • Alderi E. Araujo
  • Gilvan Ferreira da Silva
  • Josivanda P.G. Gouveia
  • Everaldo P. Medeiros
Federal University of Campina Grande, Post-graduate Program in Agricultural Engineering, CEP 58.429-140, Campina Grande-PB, Brazil

 https://orcid.org/0000-0003-0991-2068
 Search for papers by this author
Federal University of Paraiba, Graduate Program in Agronomy, CEP 58.397-000, Areia, PB, Brazil

 https://orcid.org/0000-0002-8800-5644
 Search for papers by this author
Embrapa Algodão, CEP 58428-095, Campina Grande, PB, Brazil

 https://orcid.org/0000-0001-5159-7829
 Search for papers by this author
Embrapa Algodão, CEP 58428-095, Campina Grande, PB, Brazil

 https://orcid.org/0000-0003-0257-1539
 Search for papers by this author
Embrapa Amazônia Ocidental, CEP 69010-970, Manaus, AM, Brazil

 https://orcid.org/0000-0003-2828-8299
 Search for papers by this author
Federal University of Campina Grande, Post-graduate Program in Agricultural Engineering, CEP 58.429-140, Campina Grande-PB, Brazil

 https://orcid.org/0000-0002-2047-986X
 Search for papers by this author
 Corresponding Author
Embrapa Algodão, CEP 58428-095, Campina Grande, PB, Brazil
[email protected]
 https://orcid.org/0000-0001-7204-3040
 Search for papers by this author

Hyperspectralimaging near infrared (HSI-NIR) has the potential to be used as a non-destructive approach for the analysis of new microbiological matrices of agriculture interest. This article describes a new method for accurately and rapidly classifying the etiologicalagents <em>Colletotrichum gossypii</em> (CG) and <em>C. gossypii</em> var. <em>cephalosporioides</em> (CGC) grown in a culture medium, using scattering reflectance HSI-NIR and multivariate pattern recognition analysis. Five strains of CG and 46 strains of CGC were used. CG and CGC strains were grown on Czapek-agar medium at 25 &#x00B0;C under a 12-hour photoperiod for 15 days. Molecular identification was performed as a reference for the CG and CGC classes by polymerase chainreaction of the intergenic spacer region of rDNA. The scattering coefficient <em>&#x00B5;<sub>s</sub></em> and the absorption coefficient <em>&#x00B5;<sub>a</sub></em> were obtained, which resulted in a <em>&#x00B5;<sub>s</sub></em> valuefor CG of 1.37 &#x00D7; 10<sup>19</sup> and for CGC of 5.83 &#x00D7; 10<sup>&#x2013;11</sup>. These results showed that the use of the standard normal variate was no longer essential and reduced the spectral range from 1000&#x2013;2500 nmto 1000&#x2013;1381 nm. The results evidenced two type II errors for the CG 457-2 and CGC 39 samples in the soft independent modelling model of the analogy model. There were no classification errors using the algorithm of the successive projectionsfor variable selection in linear discriminant analysis (SPA-LDA). A parallel validation of the results obtained with SPA-LDA was performed using a box plot analysis with the 11 variables selected by SPA, in which there were no outliers for the HSI-NIR models.The new HSI-NIR and SPA-LDA procedures for the classification of CG and CGC etiological agents are noted for their greater analytical speed, accuracy, simplicity, lower cost and non-destructive nature.

Keywords: fungal identification, fungal taxonomy, non-destructive analysis, cotton crop, hyperspectral image


Metrics

Downloads:

242

Full-Text HTML Views:

40

Abstract Views:

782

 

Citing this paper

 Plain Text

 Endnote

 Bibtex

 RIS