Abstract
This study was carried out for rapid and noninvasive determination of the class of sorghum species by using the manifold dimensionality reduction (MDR) method and the nonlinear regression method of least squares support vector machines (LS-SVM) combing with the mid-infrared spectroscopy (MIRS) techniques. The methods of Durbin and Run test of augmented partial residual plot (APaRP) were performed to diagnose the nonlinearity of the raw spectral data. The nonlinear MDR methods of isometric feature mapping (ISOMAP), local linear embedding, laplacian eigenmaps and local tangent space alignment, as well as the linear MDR methods of principle component analysis and metric multidimensional scaling were employed to extract the feature variables. The extracted characteristic variables were utilized as the input of LS-SVM and established the relationship between the spectra and the target attributes. The mean average precision (MAP) scores and prediction accuracy were respectively used to evaluate the performance of models. The prediction results showed that the ISOMAP-LS-SVM model obtained the best classification performance, where the MAP scores and prediction accuracy were 0.947 and 92.86%, respectively. It can be concluded that the ISOMAP-LS-SVM model combined with the MIRS technique has the potential of classifying the species of sorghum in a reasonable accuracy.
Highlights
Sorghum is a diverse genus consisting of both cultivated and wild species
The manifold learning algorithms can be divided into two categories: one is the linear methods such as PCA, LDA and LPP, and the other is the nonlinear methods such as isometric mapping (ISOMAP), local linear embedding (LLE), laplacian eigenmaps (LE) and local tangent space alignment (LTSA)
The results indicated that the nonlinear manifold dimensionality reduction (MDR) method combined with least squares support vector machines (LS-SVM) regression method demonstrated better performance than the linear manifold dimensionality reduction (DR) methods
Summary
Sorghum is a diverse genus consisting of both cultivated and wild species. Most of them have considerable genetic and morphological diversity[1]. Silk sorghum resembles Johnsongrass and is treated as a dangerous weed because of its aggressive competition with crop plants for soil nutrients, water, space, and frequent toxicity to grazing stock. In contrast to these weeds in sorghum, the species of S.sudanense and S.propinguum are the cultivated pasture plants. Chen et al classified the vinegar quality according to the total acid content through the near-infrared spectroscopy techniques and the nonlinear regression methods and obtained a satisfying result[10]. Both of the linear and nonlinear manifold DR methods were conducted to approximate the high dimensional spectral data and their performances were compared
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call