Nondestructive determination of pH value in beef using visible/near-infrared spectroscopy and genetic algorithm

In order to realize nondestructive and rapid determination of beef pH stored at 4℃ during its whole shelf-life, a laboratory visible/near-infrared spectroscopy system using visible/near-infrared spectroscopy and genetic algorithm was built to collect 120 beef samples' reflectance spectra in the 400-1700nm. These samples were stored at 4℃ for 1-18days. The reflectance spectra of samples were performed with different pretreatments, such as multiplicative scatter correction (MSC), Savitzky-Golay(SG) smoothing method. The prediction model of multiple linear regression (MLR), partial least squares regression(PLSR) and least square-support vector machine(LS-SVM) were constructed for prediction of pH value in beef with full-spectrum and effective wavelengths selected by genetic algorithm(GA), respectively. The results showed that the MSC combined with SG smoothing was the best pretreatment, and the performance of models established with effective wavelengths selected by GA were better than the full-spectrum models, and the best performance was achieved by LS-SVM model, its correlation coefficient and standard deviation were 0.935 and 0.111, respectively. The prediction accuracy was improved. This study demonstrated that the LS-SVM model built by using visible/near- infrared spectroscopy with GA could nondestructively and rapidly determine pH value in beef during its whole shelf-life. This research provides a basis of further developing device for nondestructive and rapid determine pH value in beef.