High pH thresholding of beef with VNIR hyperspectral imaging. (December 2017)

• Record Type: Journal Article
• Authors: Crichton, Stuart O.J.; Kirchner, Sascha M.; Porley, Victoria; Retz, Stefanie; von Gersdorff, Gardis; Hensel, Oliver; Sturm, Barbara
• Is Part Of: Meat science. Volume 134(2017)
• Page Start: 14
• Page End: 17
• Publication Date: 2017-12
• Subjects: Beef -- Classification -- Hyperspectral imaging -- Non-invasive measurement -- pH; Meat -- Periodicals; Meat industry and trade -- Periodicals; Viande -- Périodiques; Viande -- Industrie -- Périodiques; Meat; Meat industry and trade; Periodicals; 641.36
• DOI: 10.1016/j.meatsci.2017.07.012
• Languages: English
• ISSNs: 0309-1740

Abstract

Abstract: Initial quality grading of meat is generally carried out using invasive and occasionally destructive sampling for the purposes of pH testing. Precise pH and thresholds exist to allow the classification of different statuses of meat, e.g. for detection of dry, firm, and dark (DFD) (when dealing with cattle and sheep), or pale, soft exudative meat (when dealing with pork). This paper illustrates that threshold detection for pH level in beef with different freshness levels (fresh, fresh frozen-thawed, matured, and matured frozen-thawed). Use of support vector machine (SVM) analysis allowed for the classification of beef samples with a pH above 5.9, and below 5.6, with an accuracy of 91% and 99% respectively. Biochemical and physical conditions of the meat concerning the pH are discussed. Highlights: A novel non-invasive approach for pH thresholding of beef is presented. The classification model exhibits a robust detection of beef meat with high pH even under different freshness conditions. The biochemical and physical aspects of DFD meat and in conjunction with their spectral response are thoroughly discussed.