STUDY ON GENETIC ALGORITHMS-BASED NIR WAVELENGTH SELECTION FOR DETERMINATION OF SOLUBLE SOLIDS CONTENT IN FUJI APPLES

The determination of soluble solids content (SSC) in intact apples was analyzed by means of near-infrared spectroscopy using an acousto-optic tunable filter technology. Genetic algorithms (GAs) were performed to select wavelengths within the range from 1,065 to 1,625 nm. Partial least squares regression (PLSR) model based on GAs was compared with the nonvariable selection. Furthermore, the majority of selected wavelengths that brought on that result have been analyzed with regard to the typical absorption bands of sugars and the low-noise spectral signals. With the GAs approach, the spectral points were reduced from 88 to 17, and the low relative standard deviation of prediction (RSDp) (6.02%) and high coefficient of calibration (0.914) were achieved. The results showed that PLSR combined with GAs not only simplified and optimized the calibration model, but also improved the prediction effect of the calibration model. It was concluded that the GAs-based method extracted relevant SSC information from special spectral regions and left out the useless spectral points simultaneously. PRACTICAL APPLICATIONS The determination of soluble solids content (SSC) in intact apples is quite important for assessing the internal quality of fruit. However, the portable spectrophotometer did not provide a reliable spectral signal. The multivariate modeling of the entire spectra not only computes complexly, but also increases spectral interferences. In the present study, an acousto-optic tunable filter spectrometer was used for analysis. Genetic algorithms (GAs) were presented for the wavelength selection in order to predict the SSC in intact apples, based on partial least squares regression of near-infrared spectral data. The results demonstrated that the wavelength selection based on GAs can build a simple model and can improve the performance of the model. It was also shown that the GAs-based method extracted relevant information from special spectral regions and left out the useless spectral points simultaneously. It is valuable to establish a nondestructive measurement for developing online sensing systems.

[1]  Marcelo Blanco,et al.  NIR spectroscopy: a rapid-response analytical tool , 2002 .

[2]  Edward V. Thomas,et al.  A primer on multivariate calibration , 1994 .

[3]  R. Leardi,et al.  Variable selection for multivariate calibration using a genetic algorithm: prediction of additive concentrations in polymer films from Fourier transform-infrared spectral data , 2002 .

[4]  R. Boggia,et al.  Genetic algorithms as a strategy for feature selection , 1992 .

[5]  M A Arnold,et al.  Genetic algorithm-based wavelength selection for the near-infrared determination of glucose in biological matrixes: initialization strategies and effects of spectral resolution. , 1998, Analytical chemistry.

[6]  J. Roger,et al.  Application of LS-SVM to non-linear phenomena in NIR spectroscopy: development of a robust and portable sensor for acidity prediction in grapes , 2004 .

[7]  P. Geladi,et al.  Linearization and Scatter-Correction for Near-Infrared Reflectance Spectra of Meat , 1985 .

[8]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[9]  B. Nicolai,et al.  NON-DESTRUCTIVE MEASUREMENT OF ACIDITY, SOLUBLE SOLIDS, AND FIRMNESS OF JONAGOLD APPLES USING NIR-SPECTROSCOPY , 1998 .

[10]  Riccardo Leardi,et al.  Genetic Algorithms as a Tool for Wavelength Selection in Multivariate Calibration , 1995 .

[11]  M A Arnold,et al.  Genetic algorithm-based method for selecting wavelengths and model size for use with partial least-squares regression: application to near-infrared spectroscopy. , 1996, Analytical chemistry.

[12]  S. Kawano,et al.  Performance of a Portable near Infrared Instrument for Brix Value Determination of Intact Mango Fruit , 2003 .

[13]  P. Geladi,et al.  Selection of near Infrared Wavelengths Using Genetic Algorithms for the Determination of Seed Moisture Content , 2003 .

[14]  Fujitoshi Shinoki,et al.  Development of a Portable near Infrared Sugar-Measuring Instrument , 2002 .

[15]  C. Boschetti,et al.  A New Genetic Algorithm Applied to the near Infrared Analysis of Gasolines , 2004 .

[16]  D. T. Rossi,et al.  Effects of wavelength range on the simultaneous quantitation of polynuclear aromatic hydrocarbons with absorption spectra , 1985 .

[17]  Ingo Truppel,et al.  Spectral Measurements on ‘Elstar’ Apples during Fruit Development on the Tree , 2005 .

[18]  Chris W. Brown,et al.  Matrix representations and criteria for selecting analytical wavelengths for multicomponent spectroscopic analysis , 1982 .

[19]  R. Poppi,et al.  Determination of triglycerides in human plasma using near-infrared spectroscopy and multivariate calibration methods , 2001 .

[20]  C. B. Lucasius,et al.  Genetic algorithms in wavelength selection: a comparative study , 1994 .

[21]  Alejandro C. Olivieri,et al.  Wavelength Selection for Multivariate Calibration Using a Genetic Algorithm: A Novel Initialization Strategy , 2002, J. Chem. Inf. Comput. Sci..