Least square support vector machine for citrus greening by use of near infrared spectroscopy

Yande Liu; Huaichun Xiao; Xudong Sun; Rubing Han; Lingyu Ye; Deli Liu

doi:10.1117/12.2268792

28 February 2017 Least square support vector machine for citrus greening by use of near infrared spectroscopy

Yande Liu, Huaichun Xiao, Xudong Sun, Rubing Han, Lingyu Ye, Deli Liu

Proceedings Volume 10256, Second International Conference on Photonics and Optical Engineering; 102560D (2017) https://doi.org/10.1117/12.2268792
Event: Second International Conference on Photonics and Optical Engineering, 2016, Xi'an, China

Abstract

Citrus greening or Huanglongbing (HLB) is one of most serious citrus diseases in the world. Once a tree is infected, there is no cure. The feasibility was investigated for discriminating citrus greening by use of near infrared (NIR) spectroscopy and least square support vector machine (LS-SVM). The spectra of sound and citrus greening samples were recorded in the wavenumber range of 4000-9000 cm^-1. The preprocessing method of second derivative with a gap of seven was adapted to eliminate spectral baseline. The spectral variables were optimized by principal component analysis (PCA) and (UVE) algorithms. The unknown samples were used to access the performance of the models. Compared to the PLS-DA model, the LS-SVM was better with the input vector of the first 15 principal components and linear kernel function. The regularization factor (γ) of linear kernel function was 1.8756, and the operation time of LS-SVM model was 0.86s. The recognition error of the LS-SVM model was zero. The results showed that the combination of LS-SVM and NIR spectroscopy could detect citrus greening nondestructively and rapidly.

Citation Download Citation

Yande Liu, Huaichun Xiao, Xudong Sun, Rubing Han, Lingyu Ye, and Deli Liu "Least square support vector machine for citrus greening by use of near infrared spectroscopy", Proc. SPIE 10256, Second International Conference on Photonics and Optical Engineering, 102560D (28 February 2017); https://doi.org/10.1117/12.2268792

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available