The estimation of nitrogen status non-destructively in rice was performed using canopy spectral reflectance with visible and near-infrared reflectance (Vis/NIR) spectroscopy. The canopy spectral reflectance of rice grown with different levels of nitrogen inputs was determined at several important growth stages. This study was conducted at the experiment farm of Zhejiang University, Hangzhou, China. The soil plant analysis development (SPAD) value was used as a reference data that indirectly reflects nitrogen status in rice. A total of 64 rice samples were used for Vis/NIR spectroscopy at 325-1075 nm using a field spectroradiometer, and chemometrics of partial least square (PLS) was used for regression. The correlation coefficient (r), root mean square error of prediction, and bias in prediction set by PLS were, respectively, 0.8545, 0.7628, and 0.0521 for SPAD value prediction in tillering stage, 0.9082, 0.4452, and −0.0109 in booting stage, and 0.8632, 0.7469, and 0.0324 in heading stage. Least squares support vector machine (LS-SVM) model was compared with PLS and back propagation neural network methods. The results showed that LS-SVM was superior to the conventional linear and non-linear methods in predicting SPAD values of rice. Independent component analysis was executed to select several sensitive wavelengths (SWs) based on loading weights; the optimal LS-SVM model was achieved with SWs of 560, 575-580, 700, 730, and 740 nm for SPAD value prediction in booting stage. It is concluded that Vis/NIR spectroscopy combined with LS-SVM regression method is a promising technique to monitor nitrogen status in rice.
Due to the advances in hyperspectral sensor technology, hyperspectral images have gained great attention in precision agriculture. In practical applications, vegetation classification is usually required to be conducted first and then the vegetation of interest is discriminated from the others. This study proposes an integrated scheme (SpeSpaVS_ClassPair_ScatterMatrix) for vegetation classification by simultaneously exploiting image spectral and spatial information to improve vegetation classification accuracy. In the scheme, spectral features are selected by the proposed scatter-matrix-based feature selection method (ClassPair_ScatterMatrix). In this method, the scatter-matrix-based class separability measure is calculated for each pair of classes and then averaged as final selection criterion to alleviate the problem of mutual redundancy among the selected features, based on the conventional scatter-matrix-based class separability measure (AllClass_ScatterMatrix). The feature subset search is performed by the sequential floating forward search method. Considering the high spectral similarity among different green vegetation types, Gabor features are extracted from the top two principal components to provide complementary spatial features for spectral features. The spectral features and Gabor features are stacked into a feature vector and then the ClassPair_ScatterMatrix method is used on the formed vector to overcome the over-dimensionality problem and select discriminative features for vegetation classification. The final features are fed into support vector machine classifier for classification. To verify whether the ClassPair_ScatterMatrix method could well avoid selecting mutually redundant features, the mean square correlation coefficients were calculated for the ClassPair_ScatterMatrix method and AllClass_ScatterMatrix method. The experiments were conducted on a widely used agricultural hyperspectral image. The experimental results showed that (1) the The proposed ClassPair_ScatterMatrix method could better alleviate the problem of selecting mutually redundant features, compared to the AllClass_ScatterMatrix method; (2) compared with the representative mutual information-based feature selection methods, the scatter-matrix-based feature selection methods generally achieved higher classification accuracies, and the ClassPair_ScatterMatrix method especially, produced the highest classification accuracies with respect to both data sets (87.2% and 90.1%); and (3) the proposed integrated scheme produced higher classification accuracy, compared with the decision fusion of spectral and spatial features and the methods only involving spectral
Wheat industry is an important constituent of Northern China's overall agricultural economy. Proper disease detection using computer vision and pattern recognition has being investigated to minimize the loss, and finally achieve intelligent healthy farming. This paper proposes a new strategy of Multi-Classifier System based on SVM (support vector machine) for pattern recognition of wheat leaf diseases for higher recognition accuracy. Diseased leaf samples with Powdery Mildew, Rust Puccinia Triticina, Leaf Blight, Puccinia Striiformis were collected in the field and images were captured before a uniform black background. Three feature sets including color feature set, shape feature set and texture feature set were created for classification analysis. The proposed combination strategy was based on stacked generalization and included twolevel structure: base-level was a module of three kinds of SVM-based classifiers trained by three feature sets and meta-level was one module of SVM-based decision classifier trained by meta-feature set which are generated through a new data fusion mechanism. Compared with other single classifiers and other strategy of classifier ensembles for wheat leaf diseases, this approach is more flexible and has higher success rate of recognition.
Epigallocatechin-3-gallate (EGCG) is credited with the majority of the health benefits associated with green tea consumption. It has a high economic and medicinal value. The feasibility of using different variable selection approaches in Fourier transform near-infrared (FT-NIR) spectroscopy for a rapid and conclusive quantitative determination of EGCG in green tea was investigated. Graphically oriented multivariate calibration modeling procedures such as interval partial least squares (iPLS), synergy interval partial least squares (siPLS), and genetic algorithm optimization combined with siPLS (siPLS-GA) were applied to select the most efficient spectral variables that provided the lowest prediction error. The performance of the final model was evaluated according to the root mean square error of prediction (RMSEP) and coefficient of determination (R(2)) for the prediction set. Experimental results showed that the siPLS-GA model obtained the best results in comparison to other models. The optimal models were achieved with R(2)(p) = 0.97 and RMSEP = 0.32. The model can be obtained with only 36 variables retained and it provides a robust model with good estimation accuracy. This demonstrates the potential of NIR spectroscopy with multivariate calibration methods to quickly detect the bioactive component in green tea.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.