Spoilage Detection of Chilled Meat during Shelf Life by using Hyperspectral Imaging Technique

Abstract. Safety quality of pork meat is desired by costumers, who require freshness of meat to be maintained during shelf life in storage. Among different reasons, microorganisms are the most common cause for the spoilage of chilled pork during shelf life. Pseudomonas is one of the dominant bacteria species which multiplies during storage at 4 o C. The research was focused to develop optical technology based on hyperspectral imaging technology to detect spoilage of chilled pork meat rapidly and non-invasively. Fifty four pork meat samples were bought from local market, vacuum packed and stored in refrigerator at 4 o C. Hyperspectral scattering image in the range of 400nm to 1100nm were collected from 3 different samples on each day. After the collection of hyperspectral data, the microbiological characteristics such as Total Viable Counts (TVC) and Pseudomonas spp. were determined by standard method as reference value. The scattering profiles were fitted by four-parameter Gompertz function.Based on the hyperspectral data and reference value, prediction model was developed using support vector machines (SVM) regression. Moreover, a single prediction model for pork shelf life was established by a combination of TVC and growth of Pseudomonas spp. The model thus develop yielded regression coefficient of 0.93 for pork spoilage. The result obtained in the research demonstrates that the hyperspectral technology is precise and reliable method for rapid, non-invasive, real time detection of chilled pork quality during its shelf life.