[Progress in quality analysis of honey by infrared spectroscopy].

The detection of the quality of honey and the differentiation of adulteration are very important for quality and safety assurance. Traditionally used chemical methods were expensive and complicated, therefore they are not suitable for the requirement of wide-scale detection. In the past decade, the detection technology of honey developed with a trend of fast and high throughput detection. Spectroscopy has the fast and non-contact characteristic, and was widely used in petrifaction. This technology also has the potential for application in honey analysis. In the present study, the progress in quantitative and qualitative analysis of honey by near infrared spectroscopy (NIR) and mid infrared spectroscopy (MIR) is reviewed. The application of this two spectroscopy methods to honey detection refers to several aspects, such as quality control analysis, determination of botanical origin, determination of geographical origin and detection of adulteration. The detailed information of the detection of honey by NIR and MIR spectroscopy was analyzed, containing detection principle, technology path, accuracy, influence factors, and the development trend.

[Analysis of NIR characteristic wavelengths for apple flesh firmness based on GA and iPLS].

In the present study, the fruit flesh firmness of apple was analyzed by near infrared (NIR) spectroscopy using an FT-NIR spectrometer. The sensitive spectral regions that provide the lowest prediction error were analyzed by different well-known variable selection methods, including dynamic backward interval partial least-squares (dynamic biPLS), sequential application of backward interval partial least-squares and genetic algorithm(dynamic biPLS & GA-PLS), and iterative genetic algorithm partial least-squares (iterative GA-PLS). Iterative GA-PLS, dynamic biPLS & GA-PLS led to a distinct reduction in the number of spectral data points with better predictive quality. Furthermore, the majority of selected wavelengths were content with the characteristic of the sorption bands of fruit flesh firmness. Pectin constituents, complex non-starch polysaccharides, which are related to texture change in apple, play an important role in their harvest maturity, ripening and storage. Comparing NIR characteristic wavelengths of apple flesh firmness and typical absorption bands for pectin, it was found that characteristic wavelengths of apple flesh firmness were consistent with the pectins relevant spectral regions. Therefore, the NIR characteristic wavelengths of apple firmness based on GA and iPLS reflected the chemical component of apple and the results were reasonable.

[Using GA-dOSC method to eliminate interference of peel with prediction of apple firmness based on near infrared diffuse reflection spectra].

In the present work, "Fuji" apples from Shandong Yantai were used to take the diffuse reflection spectra by FT-NIR PLS components (i.e., factors) were computed by nonlinear iterative partial least squares (NIPALS) and the number of latent factors (LV) was optimized by a leave-one-out cross-validation procedure on the calibration set. On the basis of partial least square (PLS) regression, the models for apples' firmness before and after peeling were compared. In order to eliminate the effect of apple peel on prediction, spectral pretreatments such as multiplicative scatter correction (MSC), derivative, direct orthogonal signal correction (DOSC) and wavelengths selection based on genetic algorithms (GA) were used. Finally, the results of different spectral treatments were compared. In conclusion, the RSDp of models for apples before and after peeling was 16.71% and 12.36%, respectively, suggesting that the apple peel played a negative role in constructing good predictive models. Moreover, the traditional spectral pretreatments (such as MSC, derivative) can hardly resolve the problem. In this research, GA-DOSC played an important role in reducing the interference of apple peel. It not only reduced the wavelength variables from 1480 to 36, but also reduced the latent variables from 5 to 1. The correlation coefficient (r) was improved from 0.753 to 0.805, and the RMSECV and RMESP were reduced from 1.019 kgf x cm(-2) and 1.197 kgf x cm(-2) to 0.919 kgf x cm(-2) and 0.924 kgf x cm(-2), respectively. Especially, the RSDp was decreased remarkably from 16.71% to 12.89%. The performance of the model after GA-DOSC treatment was similar to the model using spectra of apple flesh (12.36%). It was concluded that the prediction precision based on GA-DOSC satisfied the requirement of NIR non-destruction determination of apples firmness.

[Possibilities of near-infrared spectroscopy for the assessment of principle components in honey].

The potential of near infrared spectroscopy (NIR) as a nondestructive method for determining the principle components of honeys was studied for 153 unifloral honeys and multifloral honey samples. Fourier transform near-infrared spectroscopy (FT-NIR), CCD near-infrared spectroscopy and PDA near-infrared spectroscopy were evaluated to quantitatively determine water content, fructose content and glucose content in honey. On the basis of partial-least square (PLS) regression, the models of honey were compared. The best calibration model gives the correlation coefficients of 0.978 5, 0.931 1 and 0.90 7 for water, fructose and glucose, respectively, with the root mean square error of prediction (RMSEP) of 0.410 8(%), 1.914 48(%) and 2.531 9(%) respectively. The results demonstrated that near-infrared spectrometry is a valuable, rapid and nondestructive tool for the quantitative analysis of the principle components in honey.

[Improving apple fruit quality predictions by effective correction of Vis-NIR laser diffuse reflecting images].

In the present study, improved laser-induced light backscattering imaging was studied regarding its potential for analyzing apple SSC and fruit flesh firmness. Images of the diffuse reflection of light on the fruit surface were obtained from Fuji apples using laser diodes emitting at five wavelength bands (680, 780, 880, 940 and 980 nm). Image processing algorithms were tested to correct for dissimilar equator and shape of fruit, and partial least squares (PLS) regression analysis was applied to calibrate on the fruit quality parameter. In comparison to the calibration based on corrected frequency with the models built by raw data, the former improved r from 0. 78 to 0.80 and from 0.87 to 0.89 for predicting SSC and firmness, respectively. Comparing models based on mean value of intensities with results obtained by frequency of intensities, the latter gave higher performance for predicting Fuji SSC and firmness. Comparing calibration for predicting SSC based on the corrected frequency of intensities and the results obtained from raw data set, the former improved root mean of standard error of prediction (RMSEP) from 1.28 degrees to 0.84 degrees Brix. On the other hand, in comparison to models for analyzing flesh firmness built by means of corrected frequency of intensities with the calibrations based on raw data, the former gave the improvement in RMSEP from 8.23 to 6.17 N x cm(-2).

The performance of nu-support vector regression on determination of soluble solids content of apple by acousto-optic tunable filter near-infrared spectroscopy.

The nu-support vector regression (nu-SVR) was used to construct the calibration model between soluble solids content (SSC) of apples and acousto-optic tunable filter near-infrared (AOTF-NIR) spectra. The performance of nu-SVR was compared with the partial least square regression (PLSR) and the back-propagation artificial neural networks (BP-ANN). The influence of SVR parameters on the predictive ability of model was investigated. The results indicated that the parameter nu had a rather wide optimal area (between 0.35 and 1 for the apple data). Therefore, we could determine the value of nu beforehand and focus on the selection of other SVR parameters. For analyzing SSC of apple, nu-SVR was superior to PLSR and BP-ANN, especially in the case of fewer samples and treating the noise polluted spectra. Proper spectra pretreatment methods, such as scaling, mean center, standard normal variate (SNV) and the wavelength selection methods (stepwise multiple linear regression and genetic algorithm with PLS as its objective function), could improve the quality of nu-SVR model greatly.

Study of wavelet denoising in apple's charge-coupled device near-infrared spectroscopy.

Discrete wavelet transform was used to eliminate the noise in the charge-coupled device near-infrared (CCD-NIR) spectra of apple. The influence of three parameters (wavelet function, decomposition level, and threshold) on the predictive ability of the calibration model was investigated. The result showed that the db, sym, and bior wavelet families performed well, while the coif, dmey, and haar wavelets were not able to denoise effectively. The best decomposition level was 2. The threshold selection rules of the default, Birge-Massart, and Penalty had good denoising results, while SURE, Sqtwolog, Heuristic SURE, and Minimax set all detailed coefficients to zero due to their high threshold values. The best denoising result was obtained with the combination of the bior3.3 wavelet function, two levels of decomposition, default threshold selection rule, and the soft thresholding method. The optimal model of soluble solids content was constructed. The relative standard deviation of prediction decreased from 7.79 to 5.82% after wavelet denoising.