Abstract:With the improvement of people’s living standard and the rapid development of modern food industry, food quality and safety issues attracted increasing attention. The consumption of chilled mutton showed a trend of continuous growth in recent years. Chilled mutton has become the main sales category of lamb products in China. and it is important to ensure its quality and safety. Texture profile analysis (TPA) is an important detection method for the quality and safety of chilled mutton, but there is a large data error in some TPA index values obtained during the actual measurement process. Based on the analysis of the specific measurement process of the TPA texture detection of chilled mutton, an optimizing method of TPA index extraction of chilled mutton was proposed. The actual force data was processed by the moving average filtering method, and on the basis of the data visualization method, the location of the characteristic nodes of the TPA curve was determined by constructing the TPA indexes extract optimization model. Based on the optimization method, the self-designed chilled mutton TPA detection system showed the data stability advantages, compared with the TA-XT PLUS tester，the average measurement error of adhesiveness was decreased by 12.4 percentage points. And all TPA indexes of the self-designed system showed a higher correlation with total bacteria, the absolute values of the correlation coefficient were over 0.8，which can better reflect the texture characteristics change of chilled mutton. The research result can provide more comprehensive theoretical support and practical experience for the TPA of chilled mutton, which can further improve the quality and safety of chilled mutton, so as to protect the rights of consumers.
