Detection of fish fillet substitution and mislabeling using multimode hyperspectral imaging techniques

Authors: Qin, Jianwei - Tony Qin; VASEFI, FARTASH - Collaborator; HELLBERG, ROSALEE - Chapman University; AKHBARDESH, ALIREZA - Collaborator; ISSACS, RACHEL - Chapman University; YILMAZ, AYSE - Chapman University; HWANG, CHANSONG - Us Forest Service (FS); BAEK, INSUCK - Orise Fellow; Schmidt, Walter; Kim, Moon

Submitted to: Food Control
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/11/2020
Publication Date: 3/12/2020
Citation: Qin, J., Vasefi, F., Hellberg, R.S., Akhbardesh, A., Issacs, R.B., Yilmaz, A., Hwang, C., Baek, I., Schmidt, W.F., Kim, M.S. 2020. Detection of fish fillet substitution and mislabeling using multimode hyperspectral imaging techniques. Food Control. 114:107234. https://doi.org/10.1016/j.foodcont.2020.107234.
DOI: https://doi.org/10.1016/j.foodcont.2020.107234

Interpretive Summary: A recent survey by the nonprofit organization Oceana found that 21% of fish sold in the United States was mislabeled (e.g., species, fresh/frozen-thawed, origins, and farmed-raised/wild-caught). Fish fillet is a vulnerable target of mislabeling. Mixing inexpensive species into high-priced species and substituting frozen-thawed fillets for fresh ones are two major fish fraudulent practices in seafood industry. This study developed multimode hyperspectral imaging techniques to inspect substitution and mislabeling for fish fillets. Four types of spectra (i.e., reflectance in visible and near-infrared region, fluorescence, reflectance in short-wave infrared region, and Raman) were extracted from hyperspectral images collected from fish fillets of different species and freshness conditions. Machine learning classifiers were used for fish species and freshness classifications. The reflectance spectroscopy in visible and near-infrared region demonstrated its potential to simultaneously inspect the fish species and freshness. The technique can be adopted to develop a low-cost point spectroscopy device for real-time detection of fish substitution and mislabeling, which can be used by regulatory agencies and seafood industry to authenticate fish and other seafood products.

Technical Abstract: Substitution of high-priced fish species with inexpensive alternatives and mislabeling frozen-thawed fish fillets as fresh are two important fraudulent practices of concern in the seafood industry. This study aimed to develop multimode hyperspectral imaging techniques to detect substitution and mislabeling of fish fillets. Line-scan hyperspectral images were acquired from fish fillets in four modes, including reflectance in visible and near-infrared (VNIR) region, fluorescence by 365 nm UV excitation, reflectance in short-wave infrared (SWIR) region, and Raman by 785 nm laser excitation. Fish fillets of six species (i.e., red snapper, vermilion snapper, Malabar snapper, summer flounder, white bass, and tilapia) were used for species differentiation and frozen-thawed red snapper fillets were used for freshness evaluation. All fillet samples were DNA tested to authenticate the species. A total of 24 machine learning classifiers in six categories (i.e., 25 decision tree, discriminant analysis, Naive Bayes, support vector machine, k-nearest neighbor, and ensemble) were used for fish species and freshness classifications using four types of spectral data in three different subsets (i.e., full spectra, first ten components of principal component analysis, and bands selected by a sequential feature selection method). The highest accuracies were achieved at 100% using full VNIR reflectance spectra for the species classification and 99.9% using full SWIR reflectance spectra for the freshness classification. The VNIR reflectance mode gave an overall best performance for both species and freshness inspection, and it will be further investigated as a rapid technique for detection of fish fillet substitution and mislabeling problems.