How to Resolve the Maximum Valuable Information in Complex NIR Signal: A Practicable Method Based on Wavelet Transform

A key problem in the field of near infrared (NIR) spectrum study is to obtain the valuable information from the complex NIR signal. A maximum information extraction method based on Wavelet Transform (WT) is proposed in this paper for helping the relative researchers to resolve the signal. The results show that the method can serve as an effective tool for obtaining the maximum valuable information in NIR study.

Quantitative Analysis of Soil Total Nitrogen Using Hyperspectral Imaging Technology with Extreme Learning Machine

Soil nutrient detection is important for precise fertilization. A total of 150 soil samples were picked from Lishui City. In this work, the total nitrogen (TN) content in soil samples was detected in the spectral range of 900-1700 nm using a hyperspectral imaging (HSI) system. Characteristic wavelengths were extracted using uninformative variable elimination (UVE) and the successive projections algorithm (SPA), separately. Partial least squares (PLS) and extreme learning machine (ELM) were used to establish the calibration models with full spectra and characteristic wavelengths, respectively. The results indicated that the prediction effect of the nonlinear ELM model was superior to the linear PLS model. In addition, the models using the characteristic wavelengths could also achieve good results, and the UVE-ELM model performed better, having a correlation coefficient of prediction (r_p), root-mean-square error of prediction (RMSEP), and residual prediction deviation (RPD) of 0.9408, 0.0075, and 2.97, respectively. The UVE-ELM model was then used to estimate the TN content in the soil sample and obtain a distribution map. The research results indicate that HSI can be used for the detection and visualization of the distribution of TN content in soil, providing a basis for future large-scale monitoring of soil nutrient distribution and rational fertilization.

Short communication: Selecting the most informative mid-infrared spectra wavenumbers to improve the accuracy of prediction models for detailed milk protein content

The objective of this study was to investigate the ability of mid-infrared spectroscopy (MIRS) to predict protein fraction contents of bovine milk samples by applying uninformative variable elimination (UVE) procedure to select the most informative wavenumber variables before partial least squares (PLS) analysis. Reference values (n=114) of protein fractions were measured using reversed-phase HPLC and spectra were acquired through MilkoScan FT6000 (Foss Electric A/S, Hillerød, Denmark). Prediction models were built using the full data set and tested with a leave-one-out cross-validation. Compared with MIRS models developed using standard PLS, the UVE procedure reduced the number of wavenumber variables to be analyzed through PLS regression and improved the accuracy of prediction by 6.0 to 66.7%. Good predictions were obtained for total protein, total casein (CN), and α-CN, which included αS1- and αS2-CN; moderately accurate predictions were observed for κ-CN and total whey protein; and unsatisfactory results were obtained for β-CN, α-lactalbumin, and β-lactoglobulin. Results indicated that UVE combined with PLS is a valid approach to enhance the accuracy of MIRS prediction models for milk protein fractions.

A non-destructive distinctive method for discrimination of automobile lubricant variety by visible and short-wave infrared spectroscopy

A novel method which is a combination of wavelet packet transform (WPT), uninformative variable elimination by partial least squares (UVE-PLS) and simulated annealing (SA) to extract best variance information among different varieties of lubricants is presented. A total of 180 samples (60 for each variety) were characterized on the basis of visible and short-wave infrared spectroscopy (VIS-SWNIR), and 90 samples (30 for each variety) were randomly selected for the calibration set, whereas, the remaining 90 samples (30 for each variety) were used for the validation set. The spectral data was split into different frequency bands by WPT, and different frequency bands were obtained. SA was employed to look for the best variance band (BVB) among different varieties of lubricants. In order to improve prediction precision further, BVB was processed by UVE-PLS and the optimal cutoff threshold of UVE was found by SA. Finally, five variables were mined, and were set as inputs for a least square-support vector machine (LS-SVM) to build the recognition model. An optimal model with a correlation coefficient (R) of 0.9850 and root mean square error of prediction (RMSEP) of 0.0827 was obtained. The overall results indicated that the method of combining WPT, UVE-PLS and SA was a powerful way to select diagnostic information for discrimination among different varieties of lubricating oil, furthermore, a more parsimonious and efficient LS-SVM model could be obtained.