Comparative Study on the Application of Visible and Near Infrared Hyperspectral Imaging for Fusarium Disease Assessment of Corn Seeds

Authors: , , ,

Abstract: A comparison between the classification of healthy and Fusarium Moniliforme diseased corn seeds through color image analysis and NIR Imaging techniques is presented in the paper. Results from SVM classification with selected color feature sets (from GDA, SFFS, Fisher’s discriminant ratio together with Scatter matrices) and spectra data analysis including classification of AR models with SVM and threshold value classifiers are shown for corresponding seeds from 7 Bulgarian varieties. It can be concluded that total error rate varies widely within for classification of spectra data for all varieties (0÷65%), whereas for classification of color images it retains close values (0.7÷13.3%) both for pericarp side and for the germ side.

WSEAS Transactions on Electronics, ISSN / E-ISSN: 1109-9445 / 2415-1513, Volume 10, 2019, Art. #10

PDF
XML
Cite
×
Citation Tools
WSEAS AMA
Plamen Daskalov, Eleonora Kirilova, Violeta Mancheva, Tsvetelina Georgieva, "Comparative Study on the Application of Visible and Near Infrared Hyperspectral Imaging for Fusarium Disease Assessment of Corn Seeds," WSEAS Transactions on Electronics, vol. 10, pp. 72-80, 2019, DOI:
Plamen Daskalov, Eleonora Kirilova, Violeta Mancheva, Tsvetelina Georgieva. Comparative Study on the Application of Visible and Near Infrared Hyperspectral Imaging for Fusarium Disease Assessment of Corn Seeds. WSEAS Transactions on Electronics. 2019;10:72-80.
Copy to Clipboard
Citation copied to Clipboard

Certification

---