| Development of simple, rapid and non-invasive methods for the prediction of safety and quality in seafood will greatly assist processors to produce safe and high quality seafood products. Near-infrared technology is one of the most important analytical techniques implemented in aquatic food industries. In this study, short-wavelength near-infrared (SW-NIR: 600--1100 nm) reflectance spectroscopy was used to predict safety and quality of high value aquatic food products. The use of SW-NIR spectroscopy with a fiber optic probe permits users to collect full spectra of an intact food remotely. SW-NIR light can penetrate through fish skin and scale into muscle tissue, permitting analysis of intact whole fish, fish fillets and to measure the bulk properties of the tissues. In addition, SW-NIR spectroscopic analysis only requires a few seconds per sample.; Ready-to-eat smoked aquatic food products such as hot smoked salmon in controlled atmosphere packaging are important and expensive commercial products. These products pose food safety risks if not properly processed and stored. A key preventive measure for smoked fish is adequate salt content. Therefore, an accurate measurement of salt content in these products is critical. Bruising is a particular problem with aquatic food products, which cannot be readily detected by examining the surface features of the fish. Using non-invasive grading methods based on SW-NIR could control bruise defects of fish products during processing, thereby improving product consistency and quality. Biological tissues are highly light scattering medium. One of the major fundamental problems with quantitative SW-NIR measurement is multiple light scattering in non-homogeneous tissues. In order to quantify spectral information, it is essential to know the SW-NIR light penetration depth and path length in nonhomogeneous tissues.; The objectives of this study were to evaluate the ability of SW-NIR spectroscopy (600--1100 nm) to predict salt content in hot smoked king or Chinook (Oncorhynchus tshawytscha) and chum (O. keta) salmon fillets prepared under commercial processing conditions, to detect bruises in intact Pacific pink (O. gorbuscha) and coho salmon (O. kisutch) by partial least squares (PLS) modeling, and to characterize the light penetration depth and path length of SW-NIR spectroscopy. |
