Advertisement

Using electronic nose to recognize fish spoilage with an optimum classifier

  • Meisam Vajdi
  • Mohammad J. VaridiEmail author
  • Mehdi Varidi
  • Mohebbat Mohebbi
Original Paper
  • 10 Downloads

Abstract

For automatic, rapid, accurate and objective classification of fish freshness under cold storage an electronic nose using seven metal dioxide gas sensors was developed to detect fish volatiles. Total viable count and Total volatile base nitrogen analyses were conducted simultaneously to indicate fish quality status. By sampling fish headspace, patterns were obtained during 15 storage days. 35 appropriate odor parameters were selected from each test. Principle component analysis was applied to reduce the 35-dimensional vectors to 5-dimensional vectors and clustered samples into fresh, semi fresh and spoiled. With 5-dimensional vectors as input, multilayer perceptron neural network modeled fish spoilage based on these three classes with 96.87 percent correct rate of test data. We found that the newly introduced hyper disk models maximum margin optimum classifier yielded 100 percent correct rate that could be successfully applied in industry for the diagnosis of fish spoilage.

Graphical abstract

Keywords

Electronic nose Hyper disk models Neural network Pattern recognition Quadratic programming Real-time data acquisition 

Abbreviations

CFU

Colony forming unit

EN

Electronic nose

Exp

Exponential function

G0i(t)

Sensors baseline response, Volt

Gsi(t)

Sensor response while sampling fish headspace by carrier nitrogen, Volt

Gi(t)

Response only due to fish headspace gases, Volt

Go

Average of the conductance in the interval [0–900] s

Gs

Steady-state conductance from the average of the conductance during the last 300 s

\(\left( {{\raise0.7ex\hbox{${dG}$} \!\mathord{\left/ {\vphantom {{dG} {dt}}}\right.\kern-0pt}\!\lower0.7ex\hbox{${dt}$}}} \right)\)

Slope of the dynamic conductance from the conductance curve within the interval [300–1500] s

∆Gr

Difference between first and last conductance.

\(H_{c}^{{disk}}\)

Hyper disk of a class

HDMMMOC

Hyper disk models maximum margin optimum classifier

MLPNN

Multilayer perceptron neural network

MSE

Mean-squared error

PCA

Principal component analysis

PCB

Printed circuit board

QCQP

Quadratically constrained quadratic optimization problem

RBF

Radial basis function

rc

Radius that were calculated by solving a quadratic program

sc

Center of bounding hyper sphere

SOCP

Second order cone programming

SVM

Support vector machines

(trapezoidal)

Area below the conductance curve between 300 and 1800s calculated by the ‘trapezoidal’ integration method

TGS

Taguchi gas sensors

TVBN

Total volatile base nitrogen

TVC

Total viable count

VOCs

Volatile organic compounds

\({\alpha _i}\)

Lagrange multipliers for the center

\({\left\| {x_{i}^{T} - {x_j}} \right\|^2}\)

Squared Euclidean distance between the two vectors

\(\sigma\)

Sigma values

Notes

Acknowledgements

This work was supported at Ferdowsi University of Mashhad [Grant Number 18108]. The authors gratefully acknowledge emeritus Professor Ali Jabari Azad of Physics Department for constructive collaboration.

References

  1. 1.
    J. Domingo, A. Bocio, Falc´ o G., Llobet j, Benefits and risks of fish consumption Part I. A quantitative analysis of the intake of omega-3 fatty acids and chemical contaminants. Toxicology 230, 219–226 (2007)CrossRefGoogle Scholar
  2. 2.
    O.B. Adedeji, P.O. Okerentugba, H.C. Innocent-Adiele, I.O. Okonko, Benefits, public health hazards and risks associated with fish consumption. N. Y. Sci. J. 5(9), 33–61 (2012)Google Scholar
  3. 3.
    FAO corporate document repository. (1995). Quality and changes in fresh fish, Produced by Fisheries and Aquaculture Department. http://www.fao.org/documents/en/.htm. Accessed 02 Dec 2017
  4. 4.
    A. Berna, Metal oxide sensors for electronic noses and their application to food analysis. Sensors 10, 3882–3910 (2010).  https://doi.org/10.3390/s100403882 CrossRefGoogle Scholar
  5. 5.
    P.C. Jurs, G.A. Bakken, H.E. McClelland, Computational methods for the analysis of chemical sensor array data from volatile analytes. Chem. Rev. 100, 2649–2678 (2000)CrossRefGoogle Scholar
  6. 6.
    C.-J. Du, D.-W. Sun, Pizza sauce spread classification using colour vision and support vector machines. J. Food Eng. 66, 137–145 (2005)CrossRefGoogle Scholar
  7. 7.
    N. Hasan, N. Ejaz, W. Ejaz, H.S. Kim, Meat and fish freshness inspection system based on odor sensing. Sensors 12(11), 15542–15557 (2012)CrossRefGoogle Scholar
  8. 8.
    A. Amari, N. El Barbri, E. Llobet, N. El Bari, X. Correig, B. Bouchikhi, Monitoring the freshness of moroccan sardines with a neural-network based electronic nose. Sensors 6, 1209–1223 (2006)CrossRefGoogle Scholar
  9. 9.
    D.J. Livingstone, P.T. Manallack, Statistics using neural networks: chance effects. J. Med. Chem. 36, 1295–1297 (1993).  https://doi.org/10.1021/jm00061a023 CrossRefGoogle Scholar
  10. 10.
    H. Cevikalp, B. Triggs. (2009). Large margin classifiers based on convex class models. In International Conference on Computer Vision Workshops Google Scholar
  11. 11.
    H. Cevikalp, B. Triggs, Large hyper disk based large margin classifier. Pattern Recogn. 46, 1523–1531 (2013)CrossRefGoogle Scholar
  12. 12.
    T. Mitchell, Machine learning and data mining. Commun. ACM 42(11), 30–36 (1999)CrossRefGoogle Scholar
  13. 13.
    W. Harrigan, M. McCance (1976). Laboratory methods in food and dairy microbiology. J. Basic Microbiol. 18(3), 226–227  https://doi.org/10.1002/jobm.19780180316 Google Scholar
  14. 14.
    Y. Feng, D. Sun, Determination of total viable count (TVC) in chicken breast fillets by near infrared hyperspectral imaging and spectroscopic transforms. Talanta 105, 244–249 (2013)CrossRefGoogle Scholar
  15. 15.
    T.A. Gill, Objective analysis of seafood quality. Food Rev Int 6(4), 681–714 (1990)CrossRefGoogle Scholar
  16. 16.
    E. Márquez-Ríos, E.F. Morán-Palacio, M.E. Lugo-Sánchez, V.M. Ocano-Higuera, R. Pacheco-Aguilar, Postmortem biochemical behavior of giant squid (Dosidicus gigas) mantle muscle stored in ice and its relation with quality parameters. J. Food Sci. 72(7), 356–362 (2007)CrossRefGoogle Scholar
  17. 17.
    R.A. Calvo, M.G. Partridge, M.A. Jabri (1998). A comparative study of principal component analysis techniques, in Proceedings of the 9th Australian conference on neural networks, Brisbane, QLD Google Scholar
  18. 18.
    W. Zhao, R. Chellappa, A. Krishnaswamy (1998). Discriminant analysis of principal components for face recognition, in Proceedings of the 3rd international conference on automatic face and gesture recognition. pp. 336–341Google Scholar
  19. 19.
    J.H. Kalivas, P.M. Lang, Mathematical Analysis of Spectral Orthogonality (Marcel Dekker, New York, 1993)Google Scholar
  20. 20.
    M.B. Menhaj, Fundamentals of Neural Networks (Professor Hesabi, Tehran, 1998)Google Scholar
  21. 21.
    H. Cevikalp (2011). Large margin classifier based on hyperdisks, in International Conference on Machine Learning and Applications Google Scholar
  22. 22.
    H. Cevikalp, B. Triggs, R. Polikar. (2008). Nearest hyperdisk methods for high-dimensional classification, In ICML’08: Proceedings of the 25th international conference on Machine learning, 120–127Google Scholar
  23. 23.
    D.M.J. Tax, R.P.W. Duin, Support vector data description. Mach. Learn. 54, 45–66 (2004)CrossRefGoogle Scholar
  24. 24.
    F. Alizadeh, D. Goldfarb, Second-order cone programming. Math. Program. 95, 3–51 (2003)CrossRefGoogle Scholar
  25. 25.
    V. Vural, J.G. Dy (2004). A hierarchical method for multi-class support vector machines, in International Conference on Machine Learning.  https://doi.org/10.1145/1015330.1015427
  26. 26.
    F.R. Antoine, C.I. Wei, W.S. Otwell, C.A. Sims, R.C. Littell, A.D. Hogle, M.R. Marshall, TVB-N correlation with odor evaluation and aerobic plate count in mahi-mahi (Coryphaena hippurus). J. Food Sci. 67(9), 3210–3214 (2002)CrossRefGoogle Scholar
  27. 27.
    B.A. Seibel, P.J. Walsh, Trimethylamine oxide accumulation in marine animals: relationship to acyl glycerol storage. J. Exp. Biol. 205, 297–306 (2002)Google Scholar
  28. 28.
    J. Liston, Recent advances in the chemistry of iced fish spoilage, in Chemistry and Biochemistry of Marine Food Products, ed. by R.E. Martin, G.J. Flick, C.E. Hebard, D.R. Ward (AVI Publishing Co, Westport, Conn., 1982), pp. 27–37Google Scholar
  29. 29.
    G. Farn, G.G. Sims, Chemical indices of decomposition in tuna. ed. by Kramer DE, Liston J, Seafood Quality Determination. Proceedings of an international symposium. (Elsevier, Amsterdam, 1986) pp. 175–183Google Scholar
  30. 30.
    A. Galdikas, A. Mironas, D. Senuliene, V. Strazdiene, A. Setkus, D. Zelenin, Response time based output of metal oxide gas sensors applied to evaluation of meat freshness with neural signal analysis. Sens. Actuators B 69, 258–265 (2000)CrossRefGoogle Scholar
  31. 31.
    J. Hammond, B. Marquis, R. Michaels, B. Oickle, B. Segee, J. Vetelino, A. Bushway, M. Camire, K. Davis, A semiconducting metal-oxide array for monitoring fish freshness. Sens. Actuators B 84, 113–122 (2002)CrossRefGoogle Scholar
  32. 32.
    N. El Barbri, A. Amari, M. Vinaixa, B. Bouchikhi, X. Correig, E. Llobet, Building of a metal oxide gas sensor-based electronic nose to assess the freshness of sardines under cold storage. Sens. Actuators B 128, 235–244 (2007)CrossRefGoogle Scholar
  33. 33.
    R. Olafsson, E. Martinsdottir, G. Olafsdottir, T.I. Sigfusson, J.W. Gardner, Monitoring of fish freshness using tin dioxide sensors, in Sensors and Sensory Systems for an Electronic Nose, ed. by J. Gardner, P.N. Bartlett (Editors.; Kluwer, Dordrecht, 1992), pp. 257–272CrossRefGoogle Scholar
  34. 34.
    G. Olafsdottir, H.L. Lauzon, E. Martinsdottir, K. Kristbergsson, Influence of storage temperature on microbial spoilage characteristics of haddock fillets (Melanogrammus aeglefinus) evaluated by multivariate quality prediction. Int. J. Food Microbiol. 111, 112–125 (2006)CrossRefGoogle Scholar
  35. 35.
    A.G. Martin, B. Lewis, M. Raducanu, J. Kim (2010). An array-based sensor for seafood freshness assessment. Bull. Korean Chem. Soc. 31(11), 3084–3092CrossRefGoogle Scholar
  36. 36.
    X.-Y. Tian, Q. Cai, Y.-M. Zhang, Rapid classification of hairtail fish and pork freshness using an electronic nose based on the PCA method. Sensors 12, 260–277 (2012)CrossRefGoogle Scholar
  37. 37.
    M. O’Connell, G. Valdora, G. Peltzer, R.M. Negri, A practical approach for fish freshness determinations using a portable electronic nose. Sens. Actuators B 80, 149–154 (2001)CrossRefGoogle Scholar
  38. 38.
    N.T.T. Mai, E. Martinsdóttir, K. Sveinsdóttir, G. Olafsdóttir, S. Arason, Application of quality index method, texture measurements and electronic nose to assess the freshness of atlantic herring (Clupea harengus) Stored in Ice. Int. J. Nutr. Food Eng. 3(9), 283–289 (2009)Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Food Science and TechnologyFerdowsi University of MashhadMashhadIran

Personalised recommendations