پیش‌بینی ویژگیهای بافتی پنیر موزارلای کم چرب با استفاده از تصویربرداری فراطیفی به کمک روش‌های یادگیری ماشین

نویسندگان
مهدی کاشانی نژاد 1
محمدامان ضیائی فر 2
علیرضا سلیمانی پور 3
ناصر بهنام پور 4
1 استاد گروه مهندسی صنایع غذایی، دانشکده صنایع غذایی، دانشگاه علوم کشاورزی و منابع طبیعی گرگان
2 گروه مهندسی صنایع غذایی، دانشکده صنایع غذایی، دانشگاه علوم کشاورزی و منابع طبیعی گرگان
3 گروه مهندسی مکانیک بیوسیستم، دانشکده مهندسی آب و خاک، دانشگاه علوم کشاورزی و منابع طبیعی گرگان
4 مرکز تحقیقات مدیریت سلامت و توسعه اجتماعی، گروه آمار زیستی و اپیدمیولوژی، دانشگاه علوم پزشکی گلستان
10.22034/FSCT.21.151.13
چکیده
با تغییر در شدت عملیات مکانیکی-حرارتی متفاوت، تنوع فرمولاسیون و شرایط نگهداری، 36 نمونه پنیر موزارلا کم­چرب تهیه و سختی چسبندگی، انسجام، فنریت، حالت صمغی و قابلیت جویدن آنها توسط تجزیه و تحلیل مشخصات بافت اندازه­گیری و با استفاده از تجزیه و تحلیل تک­متغیره در قالب فاکتوریل در نرم­افزار SPSS با یکدیگر مقایسه شد. سپس تصویربرداری از همان نمونه­ها با دوربین فراطیفی در محدوده 1000-400 نانومتر با دوربین فراطیفی انجام و پس از پیش­پردازش طیف­ها و جداسازی طول موج­های مؤثر به کمک الگوریتم­های انتخاب ویژگی، مدلسازی با الگوریتم رگرسیون خطی چندگانه، رگرسیون حداقل مربعات جزئی، ماشین بردار پشتیبان با کرنل خطی، شبکه عصبی پرسپترون چندلایه، جنگل­های تصادفی و الگوریتم رأی اکثریت در نرم­افزار پایتون انجام و کارائی مدل­های ارزیابی گردید. نتایج نشان داد که با تشدید عملیات مکانیکی-حرارتی، سختی، فنریت، حالت صمغی و قابلیت جویدن و انسجام افزایش و چسبندگی کاهش پیدا کرد (05/0< P). افزودن اسید و جانشین­شونده­های چربی سبب کاهش سختی، انسجام، فنریت و قابلیت جویدن شده و حالت صمغی و چسبندگی را افزایش دادند. الگوریتم رأی اکثریت، بیشترین کارایی را در پیش­بینی سختی (878/0=R2p، 52/2606= RMSEp و 12/2=RPD) بروز داد و توانست انسجام موزارلا را با کارائی بالاتری نسبت به سایر الگوریتم­ها پیش­بینی نماید. رگرسیون خطی چندگانه در پیش­بینی چسبندگی کارائی نداشت، اما روش جنگل­های تصادفی با عملکرد بالا این ویژگی را پیش­بینی نمود (808/0=R2p، 49/56= RMSEp، 90/1=RPD). شبکه عصبی پرسپترون چندلایه با کمترین خطا، توانست فنریت (848/0= R2p 094/0= RMSEp، 12/2=RPD) و قابلیت جویدن (84/0=R2p، 21/1117= RMSEp، 96/1=RPD) موزارلا را با عملکرد مناسب پیش­بینی نماید. تمام روش­ها به جز جنگل­های تصادفی توانستند با کارائی بالا حالت صمغی را پیش­بینی کنند. در این مطالعه مشخص شد عوامل فرایند تأثیر معنی­داری بر ویژگی­های بافتی داشتند و روش تصویربرداری تصویربرداری فراطیفی یک روش جایگزین مناسب برای تخمین ویژگی­های بافتی پنیر موزارلا تشخیص داده شد.
کلیدواژه‌ها
واژگان کلیدی: موزارلا کم‌چرب
ویژگی‌های بافتی
تصویربرداری فراطیفی
یادگیری ماشین

موضوعات

عنوان مقاله English

Prediction of textural characteristics in low-fat mozzarella cheese by Hyperspectral imaging using machine learning methods

نویسندگان English

Mahdi Kashaninejad 1
Aman Mohammad Ziaiifar 2
Alireza Soleimanipour 3
Naser Behnampour 4
1 Professor, Department of Food Process Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran,*Corresponding author, E-mail address: kashani@gau.ac.ir, Beheshti Ave., Gorgan. 49138-15739, Iran Tel-fax: +98(17) 32423080
2 Department of Food Process Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran
3 Department of Biosystems Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran
4 Health Management and Social Development Research Centre, Department of Biostatistics and Epidemiology, Golestan University of Medical Sciences, Gorgan, Iran
چکیده English

Changing the thermos-mechanical properties, variety of formulation and storage conditions, 36 samples of low-fat mozzarella cheese were produced and their hardness, adhesiveness, cohesiveness, springiness, cohesiveness, gumminess and chewiness were evaluated by TPA followed by analyzing data using completely randomized factorial design with univariate analysis through IBM SPSS Statistics. 26. Then, Imaging of the same samples with a Hyperspectral camera in the range of 400-1000 nm as well as pre-processing the spectra and preferring the important wavelengths by feature selection algorithms to developed the calibration models including multiple linear regression algorithms, partial least squares regression, support vector machine with a linear kernel, multilayer perceptron neural network, random forests and majority voting algorithm was performed in Python software followed by the performance of models were evaluated. Results showed that the more increased the stretching time in hot water from 2 to 8 minutes, the more the hardness, springiness, gumminess and chewiness and cohesiveness increased, but adhesiveness was decreased. The majority vote algorithm (VOTING) revealed the highest performance in hardness prediction (R2p=0.878, RMSEp=2606.52 and RPD=2.12) and was able to predict the cohesiveness of mozzarella with higher accuracy more than other algorithms. Multiple linear regression couldn’t predict the adhesiveness properly, but random forest method with high performance predicted this feature (R2p=0.808, RMSE=56.49, RPD=1.90). The multi-layer perceptron neural network with the least error, predicted springiness (R2p = 0.848, RMSEp = 0.094, RPD = 2.12) and chewiness (R2p = 0.84, RMSEp = 1117.21, RPD = 1.96) with high accuracy. All methods except random forest were able to predict the gumminess of mozzarella with high efficiency. In this study, it was cleared that the process conditions had significant effects on the textural characteristics and the Hyperspectral imaging was found to be a suitable alternative method for estimating the textural characteristics of mozzarella cheese.

کلیدواژه‌ها English

Key words: low-fat mozzarella
Textural Characteristics
Hyperspectral imaging
Machine learning
1. Fox, P. F., McSweeney, P. L., Cogan, T. M., & Guinee, T. P. 2004. Cheese: An overview. cheese: chemistry, physics and microbiology. General Asp. 1: 1226-1232.
2. Vásquez, N., Magán, C., Oblitas, J., Chuquizuta, T., Avila-George, H., & Castro, W. 2018. Comparison between artificial neural network and partial least squares regression models for hardness modeling during the ripening process of Swiss-type cheese using spectral profiles. Journal of Food Engineering. 219: 8-15.
3. Kindstedt, P. 1993. Mozzarella and Pizza Cheese. In ‘‘Cheese: Chemistry, Physics and Microbiology, Vol. 2,’’2nd Edition (PF Fox, ed.), Chapman and Hall, London.
4. Zisu, B. and N. Shah. 2005. Low-fat mozzarella as influenced by microbial exopolysaccharides, preacidification, and whey protein concentrate. Journal of Dairy Science. 88(6): 1973-1985.
5. Fogaça, D. N. L., da Silva, W. S., & Rodrigues, L. B. 2017. Influence of compression parameters on mechanical behavior of mozzarella cheese. Journal of texture studies. 48(5): 427-432.
6. Dai, S., Jiang, F., Shah, N. P., & Corke, H. 2019. Functional and pizza bake properties of Mozzarella cheese made with konjac glucomannan as a fat replacer. Food Hydrocolloids. 92: 125-134.
7. Jahani, T. and M. Azar. 2016. Effects of Fat Replacement and homogenization on the Compositional and Sensory properties of low Fat Mozzarella. Journal of Food Processing and Preservation. 8(2): 123-142.
8. Solorza, F. J., & Bell, A. E. 1995. Effect of calcium, fat and total solids on the rheology of a model soft cheese system. International journal of dairy technology. 48(4): 133-139.
9. McMahon, D. J., Alleyne, M. C., Fife, R. L., & Oberg, C. J. 1996. Use of fat replacers in low fat Mozzarella cheese. Journal of Dairy Science. 79(11): 1911-1921.
10. Wang, H.-H. and D.-W. Sun. 2002. Melting characteristics of cheese: analysis of effects of cooking conditions using computer vision technology. Journal of Food Engineering. 51(4): 305-310.
11. Gonçalves, M. C. and H. R. Cardarelli. 2020. Effect of the stretching temperature on the texture and thermophysical properties of mozzarella cheese. Journal of Food Processing and Preservation. 44(9): e14703.
12. Yu, C. and S. Gunasekaran. 2005. Modeling of melt conveying and heat transfer in a twin-screw cheese stretcher. Journal of Food Engineering. 70(2): 245-252.
13. Mulvaney, S., Barbano, D. M., & Yun, J. J. 1997. Systems analysis of the plasticization and extrusion processing of Mozzarella cheese. Journal of Dairy Science. 80(11): 3030-3039.
14. Banville, V., Chabot, D., Power, N., Pouliot, Y., & Britten, M. 2016. Impact of thermo-mechanical treatments on composition, solids loss, microstructure, and rheological properties of pasta filata–type cheese. International Dairy Journal. 61: 155-165.
15. Merrill, R. K., Oberg, C. J., & McMahon, D. J. 1994. A method for manufacturing reduced fat Mozzarella cheese. Journal of Dairy Science. 77(7): 1783-1789.
16. Revilla, I., González-Martín, I., Hernández-Hierro, J. M., Vivar-Quintana, A., González-Pérez, C., & Lurueña-Martínez, M. A. 2009. Texture evaluation in cheeses by NIRS technology employing a fibre-optic probe. Journal of Food Engineering. 92(1): 24-28.
17. Wu, D., Sun, D. W., & He, Y. 2014. Novel non-invasive distribution measurement of texture profile analysis (TPA) in salmon fillet by using visible and near infrared hyperspectral imaging. Food Chemistry. 145: 417-426.
18. Vásquez, N., Magán, C., Oblitas, J., Chuquizuta, T., Avila-George, H., & Castro, W. 2018. Comparison between artificial neural network and partial least squares regression models for hardness modeling during the ripening process of Swiss-type cheese using spectral profiles. Journal of Food Engineering. 219: 8-15.
19. Pu, H., Yu, J., Liu, Z., Paliwal, J., & Sun, D. W. (2023b). Evaluation of the Effects of Vacuum Cooling on Moisture Contents, Colour and Texture of Mushroom (Agaricus Bisporus) Using Hyperspectral Imaging Method. Microchemical Journal: 108653.
20. Priyashantha, H., Höjer, A., Saedén, K. H., Lundh, Å., Johansson, M., Bernes, G., & Hetta, M. 2020. Use of near-infrared hyperspectral (NIR-HS) imaging to visualize and model the maturity of long-ripening hard cheeses. Journal of Food Engineering. 264: 109687.
21. Kang, R., Zhao, M., Fagan, C. C., Methven, L., Oruna-Concha, M. J., & O’Donnell, C. P. 2020. Wavelength selection for rapid identification of different particle size fractions of milk powder using hyperspectral imaging. Sensors. 20(16): 4645.
22. Downey, G., Sheehan, E., Delahunty, C., O’Callaghan, D., Guinee, T., & Howard, V. 2005. Prediction of maturity and sensory attributes of Cheddar cheese using near-infrared spectroscopy. International Dairy Journal. 15(6-9): 701-709.
23. Darnay, L., Králik, F., Oros, G., Koncz, Á., & Firtha, F. 2017. Monitoring the effect of transglutaminase in semi-hard cheese during ripening by hyperspectral imaging. Journal of Food Engineering. 196: 123-129.
24. Lei, T. and D.-W. Sun. 2019. Developments of nondestructive techniques for evaluating quality attributes of cheeses: A review. Trends in Food Science & Technology. 88: 527-542.
25. Sun, D.-W. 2010. Hyperspectral imaging for food quality analysis and control, Elsevier.
26. Khojastehnazhand, M., Khoshtaghaza, M. H., Mojaradi, B., Rezaei, M., Goodarzi, M., & Saeys, W. 2014. Comparison of visible–near infrared and short wave infrared hyperspectral imaging for the evaluation of rainbow trout freshness. Food Research International. 56: 25-34.
27. Wu, D. and D.-W. Sun. 2013. Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review—Part I: Fundamentals. Innovative Food Science & Emerging Technologies. 19: 1-14.
28. Saha, D. and A. Manickavasagan. 2021. Machine learning techniques for analysis of hyperspectral images to determine quality of food products: A review. Current Research in Food Science. 4: 28-44.
29. Ah, J., & Tagalpallewar, G. P. 2017. Functional properties of Mozzarella cheese for its end use application. Journal of food science and technology. 54(12): 3766-3778.
30. Jahani, T. and M. Azar. 2016. Effects of Fat Replacement and homogenization on the Compositional and Sensory properties of low Fat Mozzarella. Journal of Food Processing and Preservation. 8(2): 123-142.
31. Feng, R., Barjon, S., van den Berg, F. W., Lillevang, S. K., & Ahrné, L. 2021. Effect of residence time in the cooker-stretcher on mozzarella cheese composition, structure and functionality. Journal of Food Engineering. 309: 110690.
32. Giménez, P., Peralta, G. H., Batistela, M. E., George, G. A., Ale, E. C., Quintero, J. P., & Bergamini, C. V. 2023. Impact of the use of skim milk powder and adjunct cultures on the composition, yield, proteolysis, texture and melting properties of Cremoso cheese. International Dairy Journal, 140, 105595.
33. Shan, J., Zhang, Y., Liang, J., & Wang, X. 2020. Characterization of the processing conditions upon textural profile analysis (tpa) parameters of processed cheese using near-infrared hyperspectral imaging. Analytical Letters. 53(8): 1190-1203.
34. Sun, D.-W. 2010. Hyperspectral imaging for food quality analysis and control, Elsevier.
35. Monicka, S. G., Manimegalai, D., & Karthikeyan, M. (2022). Detection of microcracks in silicon solar cells using Otsu-Canny edge detection algorithm. Renewable Energy Focus, 43, 183-190.
36. Le Coënt, A. L., Rivoire, A., Briancon, S., & Lieto, J. 2005. An original image-processing technique for obtaining the mixing time: The box-counting with erosions method. Powder technology, 152(1-3), 62-71.
37. Zhang, D., Lillevang, S. K., & Shah, N. P. 2021. Influence of pre-acidification, and addition of KGM and whey protein-based fat replacers CH-4560, and YO-8075 on texture characteristics and pizza bake properties of low-fat Mozzarella cheese. LWT. 137: 110384.
38. ElMasry, G. M., & Nakauchi, S. 2016. Image analysis operations applied to hyperspectral images for non-invasive sensing of food quality–A comprehensive review. Biosystems engineering, 142, 53-82.
39. Kandpal, L. M., Lee, H., Kim, M. S., Mo, C., & Cho, B. K. 2013. Hyperspectral reflectance imaging technique for visualization of moisture distribution in cooked chicken breast. Sensors. 13(10): 13289-13300.
40. Ghamisi, P., Couceiro, M. S., Fauvel, M., & Benediktsson, J. A. 2013. Integration of segmentation techniques for classification of hyperspectral images. IEEE Geoscience and Remote Sensing Letters, 11(1), 342-346.
41. Khan, A., Munir, M. T., Yu, W., & Young, B. 2020. Wavelength selection for rapid identification of different particle size fractions of milk powder using hyperspectral imaging. Sensors, 20(16), 4645.
42. Park, B. and R. Lu. 2015. Hyperspectral imaging technology in food and agriculture, Springer.
43. Wold, S., Sjostrom, M., & Eriksson, L. 2001. PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratoary Systems. 58: 109–130.
44. Nicolai, B. M., Beullens, K., Bobelyn, E., Peirs, A., Saeys, W., Theron, K. I., & Lammertyn, J. 2007. Nondestructive measurement of fruit and vegetable quality by means of NIR spectroscopy: A review. Postharvest Biology and Technology. 46(2): 99-118.
45. Pan, T. T., Sun, D. W., Cheng, J. H., & Pu, H. (2016). Regression algorithms in hyperspectral data analysis for meat quality detection and evaluation. Comprehensive reviews in food science and food safety, 15(3), 529-541.
46. Kuhn, M. and K. Johnson. 2013. Applied predictive modeling, Springer.
47. Geron, A. 2017. Hands-On Machine Learning with Scikit-Learn & Tensorflow O’Reilly Media, Inc. 1005 Gravenstein Highway North, Sebastopol, CA 95472: 564.
48. Seo, D. K., Kim, Y. H., Eo, Y. D., Lee, M. H., & Park, W. Y. (2018). Fusion of SAR and multispectral images using random forest regression for change detection. ISPRS International Journal of Geo-Information, 7(10), 401.
49. Yang, H., Li, F., Wang, W., & Yu, K. 2021. Estimating above-ground biomass of potato using random forest and optimized hyperspectral indices. Remote Sensing, 13(12), 2339.
50. Zhou, L., Zhang, C., Liu, F., Qiu, Z., & He, Y. 2019. Application of deep learning in food: a review. Comprehensive Reviews in Food Science and Food Safety. 18(6): 1793-1811.
51. Jin, X., Xiao, Z. Y., Xiao, D. X., Dong, A., Nie, Q. X., Wang, Y. N., & Wang, L. F. 2022. Quantitative inversion model of protein and fat content in milk based on hyperspectral techniques. International Dairy Journal. 134: 105467.
52. Granato, D. and G. Ares. 2014. Mathematical and statistical methods in food science and technology, John Wiley & Sons.
53. Ashfaque, J. M. and A. Iqbal. 2019. Introduction to support vector machines and kernel methods. publication at https://www. researchgate. net/publication/332370436.
54. Esen, M. K. and N. Güzeler. 2023. The effects of the use of whey protein as a fat replacer on the composition, proteolysis, textural, meltability, microstructural, and sensory properties of reduced-fat Boru-type Künefe cheese during storage. International Dairy Journal. 137: 105519.
55. Metzger, L. E., Barbano, D. M., Kindstedt, P. S., & Guo, M. R. 2001. Effect of milk preacidification on low fat Mozzarella cheese: II. Chemical and functional properties during storage. Journal of Dairy Science. 84(6): 1348-1356.
56. Amador-Espejo, G. G., Ruiz-Lopez, I. I., Gibbens-Bandala, P. J., Delgado-Macuil, R. J., & Ruiz-Espinosa, H.2021. Thermosonicated whey protein concentrate blends on quality attributes of reduced fat Panela cheese. Ultrasonics Sonochemistry. 76: 105621.
57. Abd Elkader, R. S. Awaad, R. A., Rizk Hassan, Z. M., & Salama, W. M. 2019. Production of low-fat white soft cheese using sodium caseinate and/or butter milk powder as a fat replacer. Arab Universities Journal of Agricultural Sciences. 27(2): 1503-1511.
58. Ismail, M., AMMAR, E. T., & El‐Metwally, R. A. I. D. 2011. Improvement of low fat mozzarella cheese properties using denatured whey protein. International journal of dairy technology. 64(2): 207-217.
59. Nateghi, L., Roohinejad, S., Totosaus, A., Rahmani, A., Tajabadi, N., Meimandipour, A., & Manap, M. Y. A. 2012. Physicochemical and textural properties of reduced fat Cheddar cheese formulated with xanthan gum and/or sodium caseinate as fat replacers. J. Food Agr. Environ. 10: 59-63.
60. Kindstedt, P. S., Kiely, L. J., & Gilmore, J. A. 1992. Variation in composition and functional properties within brine-salted Mozzarella cheese. Journal of Dairy Science. 75(11): 2913-2921.
61. Tunick, M. H., Mackey, K. L., Smith, P. W., & Holsinger, V. H. 1991. Effects of composition and storage on the texture of Mozzarella cheese. Nederlands melk en Zuiveltijdschrift. 45(2): 117-125.
62. Alinovi, M., Wiking, L., Corredig, M., & Mucchetti, G. 2020. Effect of frozen and refrigerated storage on proteolysis and physicochemical properties of high-moisture citric mozzarella cheese. Journal of Dairy Science. 103(9): 7775-7790.
63. Topcu, A., Bulat, T., & Özer, B. 2020. Process design for processed Kashar cheese (a pasta-filata cheese) by means of microbial transglutaminase: Effect on physical properties, yield and proteolysis. LWT. 125: 109226.
64. Shafiee, S., Polder, G., Minaei, S., Moghadam-Charkari, N., Van Ruth, S., & Kuś, P. M. 2016. Detection of honey adulteration using hyperspectral imaging. IFAC-PapersOnLine. 49(16): 311-314.
65. Jajromi, A., Taghi Zadeh, M., Sazgar Nia, A., & Behzad, K. 2015. Application of preprocessing techniques for visible/near infrared spectroscopy data in developing a model for the prediction of soluble solid and acidity of lime. Journal of food science and technology (Iran). 13(53): 103-112.
66. Park, S., Yang, M., Yim, D. G., Jo, C., & Kim, G. 2023. Industrial freezing and tempering for optimal functional properties in thawed Mozzarella cheese. Food Chemistry. 405: 134933.
67. Zou, Z., Wu, Q., Long, T., Zou, B., Zhou, M., Wang, Y., & Xu, L. 2023. Classification and Adulteration of Mengding Mountain Green Tea Varieties Based on Fluorescence Hyperspectral Image Method. Journal of Food Composition and Analysis: 105141.
68. Rinnan, Å., Van Den Berg, F., & Engelsen, S. B. 2009. Review of the most common pre-processing techniques for near-infrared spectra. TrAC Trends in Analytical Chemistry. 28(10): 1201-1222.
69. Teixido-Orries, I., Molino, F., Femenias, A., Ramos, A. J., & Marín, S. 2023. Quantification and classification of deoxynivalenol-contaminated oat samples by near-infrared hyperspectral imaging. Food Chemistry: 135924.
70. Ni, F., Zhu, X., Gu, F., & Hu, Y. 2019. Nondestructive detection of apple crispness via optical fiber spectroscopy based on effective wavelengths. Food Science & Nutrition. 7(11): 3654-3663.
71. Sahadevan, A. S., Routray, A., Das, B. S., & Ahmad, S. 2016. Hyperspectral image preprocessing with bilateral filter for improving the classification accuracy of support vector machines. Journal of Applied Remote Sensing. 10(2): 025004-025004.
72. Zhang, D., Xu, Y., Huang, W., Tian, X., Xia, Y., Xu, L., & Fan, S. 2019. Nondestructive measurement of soluble solids content in apple using near infrared hyperspectral imaging coupled with wavelength selection algorithm. Infrared Physics & Technology. 98: 297-304.
73. Zhou, Q., Huang, W., Tian, X., Yang, Y., & Liang, D. 2021. Identification of the variety of maize seeds based on hyperspectral images coupled with convolutional neural networks and subregional voting. Journal of the Science of Food and Agriculture. 101(11): 4532-4542.
74. Zhang, J., Ma, Y., Liu, G., Fan, N., Li, Y., & Sun, Y. (2022). Rapid evaluation of texture parameters of Tan mutton using hyperspectral imaging with optimization algorithms. Food Control, 135, 108815.
75. Zhang, J., Guo, Z., Ren, Z., Wang, S., Yin, X., Zhang, D., & Ma, C. 2023. A non-destructive determination of protein content in potato flour noodles using near-infrared hyperspectral imaging technology. Infrared Physics & Technology. 130: 104595.
76. González-Martín, M. I., Hernández-Hierro, J. M., Revilla, I., Vivar-Quintana, A., González-Pérez, C., García, L. G., & Ortega, I. A. L. 2011. Differentiation of organic and non-organic ewe's cheeses using main mineral composition or near infrared spectroscopy coupled to chemometric tools: A comparative study. Talanta. 85(4): 1915-1919.