Classification of Automobile Lubricant by Near-Infrared Spectroscopy Combined with Machine Classification

Successive projections algorithm (SPA) combined with least square-support vector machine (LS-SVM) was investigated to determine the citric acid of lemon vinegar by 13 wavelengths within visible/near infrared (Vis/NIR) spectral region. Five concentration levels (100%, 80%, 60%, 40% and 20%) of lemon vinegar were prepared, and the calibration set consisted of 150 samples, validation set consisted of 75 samples and the remaining 75 samples for prediction set. After the comparison of different preprocessing such as smoothing, standard normal variate and derivative, SPA was applied to extract the effective wavelengths to reduce the redundancies and collinearity of variables, and the multiple linear regression (MLR) models were developed compared with partial least squares (PLS) models. Simultaneously, the selected wavelengths were used as the inputs of LS-SVM, and a new proposed combination of SPA-LS-SVM model was developed. The results indicated that SPA-LS-SVM achieved the optimal prediction performance, and the correlation coefficient (r) and root mean square error of prediction (RMSEP) were 0.9894 and 0.0623, respectively. An excellent prediction precision was obtained. The overall results demonstrated that it was feasible to utilize Vis/NIR spectroscopy to predict the citric acid of lemon vinegar, and SPA-LS-SVM model achieved the optimal prediction precision. This study supplied a feasible method for the process monitoring of fruit vinegar manufacture and fermentation.

Download Full-text

Spectroscopy quantitative analysis cotton content of blend fabrics

International Journal of Clothing Science and Technology ◽

10.1108/ijcst-07-2015-0076 ◽

2016 ◽

Vol 28 (1) ◽

pp. 65-76 ◽

Cited By ~ 1

Author(s):

Xudong Sun ◽

Mingxing Zhou ◽

Yize Sun

Keyword(s):

Near Infrared ◽

Multivariate Calibration ◽

Partial Least Square ◽

Least Square ◽

Support Vector ◽

Content Type ◽

Svm Model ◽

Calibration Methods ◽

Learning Machine ◽

Mean Square Errors

Purpose – The purpose of this paper is to develop near infrared (NIR) techniques coupled with multivariate calibration methods to rapid measure cotton content in blend fabrics. Design/methodology/approach – In total, 124 and 41 samples were used to calibrate models and assess the performance of the models, respectively. Multivariate calibration methods of partial least square (PLS), extreme learning machine (ELM) and least square support vector machine (LS-SVM) were employed to develop the models. Through comparing the performance of PLS, ELM and LS-SVM models with new samples, the optimal model of cotton content was obtained with LS-SVM model. The correlation coefficient of prediction (r p ) and root mean square errors of prediction were 0.98 and 4.50 percent, respectively. Findings – The results suggest that NIR technique combining with LS-SVM method has significant potential to quantitatively analyze cotton content in blend fabrics. Originality/value – It may have commercial and regulatory potential to avoid time consuming work, costly and laborious chemical analysis for cotton content in blend fabrics.

Download Full-text

Transferability and scaling of soil total carbon prediction models in Florida

10.7287/peerj.preprints.494v1 ◽

2014 ◽

Author(s):

Sabine Grunwald ◽

Congrong Yu ◽

Xiong Xiong

Keyword(s):

Prediction Error ◽

Near Infrared ◽

Prediction Models ◽

Regional Scale ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Total Carbon ◽

Support Vector ◽

Domain Iii

The applicability, transfer, and scalability of visible/near-infrared (VNIR)-derived soil models are still poorly understood. The objectives of this study in Florida, U.S. were to: (i) compare three methods to predict soil total carbon (TC) using five fields (local scale) and a pooled (regional scale) VNIR spectral dataset, (ii) assess the model’s transferability among fields, and (iii) evaluate the up- and down-scaling behavior of TC prediction models. A total of 560 TC-spectral sets were modeled by Partial Least Square Regression (PLSR), Support Vector Machine (SVM), and Random Forest. The transferability and up- and down-scaling of models were limited by the following factors: (i) the spectral data domain, (ii) soil attribute domain, (iii) methods that describe the internal model structure of VNIR-TC relationships, and (iv) environmental domain space of attributes that control soil carbon dynamics. All soil logTC models showed excellent performance based on all three methods with R2 > 0.86, bias < 0.01%, root mean square prediction error (RMSE) = 0.09%, residual predication deviation (RPD) > 2.70% , and ratio of prediction error to inter-quartile range (RPIQ) > 4.54. PLSR performed substantially better than SVM to scale and transfer models. Upscaled soil TC models performed somewhat better in terms of model fit (R2), RPD, and RPIQ, whereas downscaled models showed less bias and smaller RMSE based on PLSR. Given the many factors that can impinge on empirically derived soil spectral prediction models, as demonstrated by this study, more focus on the applicability and scaling of them is needed.

Download Full-text

Rapid Detection of Melamine in Milk Powder by near Infrared Spectroscopy

Journal of Near Infrared Spectroscopy ◽

10.1255/jnirs.829 ◽

2009 ◽

Vol 17 (2) ◽

pp. 59-67 ◽

Cited By ~ 57

Author(s):

Chenghui Lu ◽

Bingren Xiang ◽

Gang Hao ◽

Jianping Xu ◽

Zhengwu Wang ◽

...

Keyword(s):

Support Vector Machine ◽

Classification Accuracy ◽

Rapid Detection ◽

Near Infrared ◽

Milk Powder ◽

Nir Spectroscopy ◽

Accurate Method ◽

Partial Least Square ◽

Least Square ◽

Support Vector

This paper establishes a novel and rapid method for detecting pure melamine in milk powder using near infrared (NIR) spectroscopy based on least squares-support vector machine (LS-SVM). Partial least square discriminant analysis (PLS-DA) was used for the extraction of principal components (PCs). The scores of the first two PCs have been applied as inputs to LS-SVM. Compared to PLS-DA, the performance of LS-SVM was better, with higher classification accuracy, both 100% for the training and testing set. The detection limit was lower than 1 ppm. Based on the results, it was concluded that NIR spectroscopy combined with LS-SVM could be used as a rapid and accurate method for detecting pure melamine in milk powder.

Download Full-text

Deep Spectral-Spatial Features of Near Infrared Hyperspectral Images for Pixel-Wise Classification of Food Products

Sensors ◽

10.3390/s20185322 ◽

2020 ◽

Vol 20 (18) ◽

pp. 5322

Author(s):

Hongyan Zhu ◽

Aoife Gowen ◽

Hailin Feng ◽

Keping Yu ◽

Jun-Li Xu

Keyword(s):

Near Infrared ◽

Principal Component ◽

Food Products ◽

Partial Least Square ◽

Least Square ◽

Support Vector ◽

White Stripe ◽

Practical Applications ◽

Spatial Features

Hyperspectral imaging (HSI) emerges as a non-destructive and rapid analytical tool for assessing food quality, safety, and authenticity. This work aims to investigate the potential of combining the spectral and spatial features of HSI data with the aid of deep learning approach for the pixel-wise classification of food products. We applied two strategies for extracting spatial-spectral features: (1) directly applying three-dimensional convolution neural network (3-D CNN) model; (2) first performing principal component analysis (PCA) and then developing 2-D CNN model from the first few PCs. These two methods were compared in terms of efficiency and accuracy, exemplified through two case studies, i.e., classification of four sweet products and differentiation between white stripe (“myocommata”) and red muscle (“myotome”) pixels on salmon fillets. Results showed that combining spectral-spatial features significantly enhanced the overall accuracy for sweet dataset, compared to partial least square discriminant analysis (PLSDA) and support vector machine (SVM). Results also demonstrated that spectral pre-processing techniques prior to CNN model development can enhance the classification performance. This work will open the door for more research in the area of practical applications in food industry.

Download Full-text

Transferability and scaling of soil total carbon prediction models in Florida

10.7287/peerj.preprints.494 ◽

2014 ◽

Author(s):

Sabine Grunwald ◽

Congrong Yu ◽

Xiong Xiong

Keyword(s):

Prediction Error ◽

Near Infrared ◽

Prediction Models ◽

Regional Scale ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Total Carbon ◽

Support Vector ◽

Domain Iii

The applicability, transfer, and scalability of visible/near-infrared (VNIR)-derived soil models are still poorly understood. The objectives of this study in Florida, U.S. were to: (i) compare three methods to predict soil total carbon (TC) using five fields (local scale) and a pooled (regional scale) VNIR spectral dataset, (ii) assess the model’s transferability among fields, and (iii) evaluate the up- and down-scaling behavior of TC prediction models. A total of 560 TC-spectral sets were modeled by Partial Least Square Regression (PLSR), Support Vector Machine (SVM), and Random Forest. The transferability and up- and down-scaling of models were limited by the following factors: (i) the spectral data domain, (ii) soil attribute domain, (iii) methods that describe the internal model structure of VNIR-TC relationships, and (iv) environmental domain space of attributes that control soil carbon dynamics. All soil logTC models showed excellent performance based on all three methods with R2 > 0.86, bias < 0.01%, root mean square prediction error (RMSE) = 0.09%, residual predication deviation (RPD) > 2.70% , and ratio of prediction error to inter-quartile range (RPIQ) > 4.54. PLSR performed substantially better than SVM to scale and transfer models. Upscaled soil TC models performed somewhat better in terms of model fit (R2), RPD, and RPIQ, whereas downscaled models showed less bias and smaller RMSE based on PLSR. Given the many factors that can impinge on empirically derived soil spectral prediction models, as demonstrated by this study, more focus on the applicability and scaling of them is needed.

Download Full-text

Rapid Determining Contents of the Rhubarb Anthraquinones Compounds by Support Vector Machines Modeling based on Near Infrared Spectra

Current Analytical Chemistry ◽

10.2174/1573411016666200317111412 ◽

2020 ◽

Vol 16 ◽

Author(s):

Linqi Liu ◽

JInhua Luo ◽

Chenxi Zhao ◽

Bingxue Zhang ◽

Wei Fan ◽

...

Keyword(s):

Infrared Spectra ◽

Near Infrared ◽

Mean Squared Error ◽

Rapid Determination ◽

Partial Least Square ◽

Least Square ◽

Support Vector ◽

Near Infrared Spectra ◽

Aloe Emodin ◽

Relative Differences

BACKGROUND: Measuring medicinal compounds to evaluate their quality and efficacy has been recognized as a useful approach in treatment. Rhubarb anthraquinones compounds (mainly including aloe-emodin, rhein, emodin, chrysophanol and physcion) are its main effective components as purgating drug. In the current Chinese Pharmacopoeia, the total anthraquinones content is designated as its quantitative quality and control index while the content of each compound has not been specified. METHODS: On the basis of forty rhubarb samples, the correlation models between the near infrared spectra and UPLC analysis data were constructed using support vector machine (SVM) and partial least square (PLS) methods according to Kennard and Stone algorithm for dividing the calibration/prediction datasets. Good models mean they have high correlation coefficients (R2) and low root mean squared error of prediction (RMSEP) values. RESULTS: The models constructed by SVM have much better performance than those by PLS methods. The SVM models have high R2 of 0.8951, 0.9738, 0.9849, 0.9779, 0.9411 and 0.9862 that correspond to aloe-emodin, rhein, emodin, chrysophanol, physcion and total anthraquinones contents, respectively. The corresponding RMSEPs are 0.3592, 0.4182, 0.4508, 0.7121, 0.8365 and 1.7910, respectively. 75% of the predicted results have relative differences being lower than 10%. As for rhein and total anthraquinones, all of the predicted results have relative differences being lower than 10%. CONCLUSION: The nonlinear models constructed by SVM showed good performances with predicted values close to the experimental values. This can perform the rapid determination of the main medicinal ingredients in rhubarb medicinal materials.

Download Full-text

Histamine Control in Raw and Processed Tuna: A Rapid Tool Based on NIR Spectroscopy

Foods ◽

10.3390/foods10040885 ◽

2021 ◽

Vol 10 (4) ◽

pp. 885

Author(s):

Sergio Ghidini ◽

Luca Maria Chiesa ◽

Sara Panseri ◽

Maria Olga Varrà ◽

Adriana Ianieri ◽

...

Keyword(s):

Root Mean Square ◽

Near Infrared ◽

Nir Spectroscopy ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Coefficient Of Determination ◽

Mean Square ◽

Tuna Fish ◽

Mean Square Errors

The present study was designed to investigate whether near infrared (NIR) spectroscopy with minimal sample processing could be a suitable technique to rapidly measure histamine levels in raw and processed tuna fish. Calibration models based on orthogonal partial least square regression (OPLSR) were built to predict histamine in the range 10–1000 mg kg−1 using the 1000–2500 nm NIR spectra of artificially-contaminated fish. The two models were then validated using a new set of naturally contaminated samples in which histamine content was determined by conventional high-performance liquid chromatography (HPLC) analysis. As for calibration results, coefficient of determination (r2) > 0.98, root mean square of estimation (RMSEE) ≤ 5 mg kg−1 and root mean square of cross-validation (RMSECV) ≤ 6 mg kg−1 were achieved. Both models were optimal also in the validation stage, showing r2 values > 0.97, root mean square errors of prediction (RMSEP) ≤ 10 mg kg−1 and relative range error (RER) ≥ 25, with better results showed by the model for processed fish. The promising results achieved suggest NIR spectroscopy as an implemental analytical solution in fish industries and markets to effectively determine histamine amounts.

Download Full-text

Evaluation of Minimum Preparation Sampling Strategies for Sugarcane Quality Prediction by vis-NIR Spectroscopy

Sensors ◽

10.3390/s21062195 ◽

2021 ◽

Vol 21 (6) ◽

pp. 2195

Author(s):

Lucas de Paula Corrêdo ◽

Leonardo Felipe Maldaner ◽

Helizani Couto Bazame ◽

José Paulo Molin

Keyword(s):

Sample Preparation ◽

Prediction Models ◽

Spectral Response ◽

Nir Spectroscopy ◽

Quality Parameters ◽

Partial Least Square ◽

Least Square ◽

Sample Type ◽

Sampling Strategies ◽

Significant Difference

Proximal sensing for assessing sugarcane quality information during harvest can be affected by various factors, including the type of sample preparation. The objective of this study was to determine the best sugarcane sample type and analyze the spectral response for the prediction of quality parameters of sugarcane from visible and near-infrared (vis-NIR) spectroscopy. The sampling and spectral data acquisition were performed during the analysis of samples by conventional methods in a sugar mill laboratory. Samples of billets were collected and four modes of scanning and sample preparation were evaluated: outer-surface (‘skin’) (SS), cross-sectional scanning (CSS), defibrated cane (DF), and raw juice (RJ) to analyze the parameters soluble solids content (Brix), saccharose (Pol), fibre, pol of cane and total recoverable sugars (TRS). Predictive models based on Partial Least Square Regression (PLSR) were built with the vis-NIR spectral measurements. There was no significant difference (p-value > 0.05) between the accuracy SS and CSS samples compared to DF and RJ samples for all prediction models. However, DF samples presented the best predictive performance values for the main sugarcane quality parameters, and required only minimal sample preparation. The results contribute to advancing the development of on-board quality monitoring in sugarcane, indicating better sampling strategies.

Download Full-text

Cancer Discrimination Using Fourier Transform Near-Infrared Spectroscopy with Chemometric Models

Journal of Chemistry ◽

10.1155/2015/619685 ◽

2015 ◽

Vol 2015 ◽

pp. 1-8 ◽

Cited By ~ 1

Author(s):

Hui Chen ◽

Zan Lin ◽

Chao Tan

Keyword(s):

Cancer Diagnosis ◽

Near Infrared ◽

Nir Spectroscopy ◽

Principal Component ◽

Classification Model ◽

Support Vector ◽

Biomedical Analysis ◽

First Derivative ◽

Svm Model ◽

High Prediction

Near-infrared (NIR) spectroscopy technique offers many potential advantages as tool for biomedical analysis since it enables the subtle biochemical signatures related to pathology to be detected and extracted. In conjunction with advanced chemometrics, NIR spectroscopy opens the possibility of their use in cancer diagnosis. The study focuses on the application of near-infrared (NIR) spectroscopy and classification models for discriminating colorectal cancer. A total of 107 surgical specimens and a corresponding NIR diffuse reflection spectral dataset were prepared. Three preprocessing methods were attempted and least-squares support vector machine (LS-SVM) was used to build a classification model. The hybrid preprocessing of first derivative and principal component analysis (PCA) resulted in the best LS-SVM model with the sensitivity and specificity of 0.96 and 0.96 for the training and 0.94 and 0.96 for test sets, respectively. The similarity performance on both subsets indicated that overfitting did not occur, assuring the robustness and reliability of the developed LS-SVM model. The area of receiver operating characteristic (ROC) curve was 0.99, demonstrating once again the high prediction power of the model. The result confirms the applicability of the combination of NIR spectroscopy, LS-SVM, PCA, and first derivative preprocessing for cancer diagnosis.

Download Full-text