Determine Internal Quality Of Strawberry By NIR Based On Wavelength Region Selection Method

Since the existing defects including much subjectivity in sensory results, wasting much time and expense in chemical analysis, detection processes can't meet the demand of rapid determination to the valid compositions of strawberry in manufacturing process and commercial process. Near infrared (NIR) reflectance spectroscopy is a fast, stable, accurate and nondestructive technique that can be employed as a replacement of time-consuming conventional physical and chemical analysis method in determination of food and agricultural produce. Strawberry spectral contains a lot of absorption peaks produced by H groups and spectra overlapping. They can affect the model accuracy and stability. In the dissertation, strawberry is the main study objective and the internal qualities criteria of strawberry are the main testing indices using NIR. The spectra model was simplified at the same time model accuracy and stability were improved. The main research work and conclusion of the dissertation are as follows:1. Conventional testing methods were used to determine internal qualities criteria data of strawberry(soluble solids content, vitamin C, acid and firmness), at the same time NIR spectra were collected by near infrared spectroscopy instrument, Standard Normal Vitiate transformation(SNV), multiplicative scatter correction(MSC), First Derivatives(1st Der), Second Derivatives(2nd Der) were used to process spectra.2. NIR spectroscopy combined with interval partial least square (iPLS) was applied to building the quantitative model to quantitatively predicts the internal qualities criteria of strawberry. In building model, the effects on the intervals number, spectral preprocessing methods and principal components factors (PCs) to results were discussed. The correlation coefficient (R) between the predicted and the reference results for the test set is used as an evaluation parameter for the models:The soluble solids content, vitamin C, acid and the firmness model R=0.9185,0.9154,0.9125 and 0.9005 respectively.. It can be concluded that internal qualities criteria of strawberry can be analyzed fast by NIR spectroscopy coupled with the appropriate chemometrics methods, and this real-time, at-site measurement will significantly improve the efficiency of quality control and assurance. 3. Selection of the efficient wavelength regions in FT-NIR spectroscopy was used for improving the model accuracy and stability. Two kinds of wavelength region selection algorithm were designed-genetic algorithm interval partial least-squares (GA-iPLS) and simulated annealing algorithm interval partial least-squares (SA-iPLS). Combination with interval partial least-squares (iPLS), synergy interval partial least-squares (siPLS) and backward interval partial least-squares (biPLS), these algorithms were used to select optimal region from SSC spectra of strawberry. The result shows that these algorithms can improve the model accuracy and simplified SSC model. The performance of SA-iPLS model was best in the four methods, and the optimal model was achieved with correlation coefficient R= 0.9451 in prediction set.4. NIR data processing soft development. Base on summarize of near infrared spectroscopy data process methods in our lab, the NIR data processing soft was developed. The soft contains calibration/prediction samples selection, pre-processing method, wavelength region selection, calibration model building and spectra prediction modules. It simplified NIR data processing and improved process efficiency.The research result shows that it is possible to use NIR spectra technical to determine internal qualities criteria of strawberry; model accuracy and stability could improved when chose proper spectra pre-processing method and.wavelength region selection method. This research is of great significant in improve the level of strawberry determination in our country.