A CHEAPLY NON-DESTRUCTIVE TECHNIQUE TO ESTIMATE HONEY QUALITY: THERMAL IMAGING AND MACHINE LEARNING

Yıl 2024, , 79 - 92, 29.05.2024

Mustafa Kibar

https://doi.org/10.31467/uluaricilik.1429971

Öz

The aim of this study was to estimate honey quality based on proline and Brix content using a thermal imaging and machine learning algorithm. The proline, Brix and color properties of twenty honey samples were determined. Proline and Brix values were classified and estimated using the classification and regression tree (CART) algorithm. The mean proline and Brix content in honey samples was 678.83±192.16 mg/kg and 83.2±0.79%, respectively. CART analysis revealed that high proline honey samples had L values above 48.143 and b* values below 35.416. In contrast, honey samples with low Brix values were characterized by L and a* values below 55.860 and 53.660, respectively, and were identified as freshly harvested. The CART algorithm classified the proline and Brix values with an accuracy of 95% and 100%, respectively (p< 0.001). As a result, whitish, bluish, blackish and greenish honeys are of higher quality due to high proline and low Brix content. However, to accurately assess honey quality based on its color traits, comprehensive studies with more honey samples and origin, are required.

Anahtar Kelimeler

Adulteration, Proline, Brix, Honey quality, CART algorithm

Bal Kalitesini Tahmin Etmek İçin Ucuz, Tahribatsiz Bir Teknik: Termal Görüntüleme ve Makine Öğrenimi

Öz

Bu çalışmanın amacı, termal görüntüleme ve makine öğrenmesi yaklaşımı kullanılarak baldaki prolin ve Brix içeriğine dayalı bal kalitesinin tahmin edilmesidir. 20 farklı bal örneğine ait prolin, Brix ve renk özellikleri belirlendi. Prolin ve Brix seviyeleri, sınıflandırma ve regresyon ağacı algoritması kullanılarak tahmin edildi ve sınıflandırıldı. Ballarda ortalama prolin ve Brix içeriği sırasıyla 678,83±192,16 mg/kg ve %83,2±0,79 olarak belirlendi. CART analizi ile yüksek prolinli balların L değerlerinin 48.143'ün üzerinde ve b* değerlerinin ise 35.416'nın altında olduğu tespit edildi. Ancak, Brix değeri düşük olan balların ise sırasıyla 55.860 ve 53.660'ın altında L ve a* değerlerine sahip olduğu ve yeni hasat edildiği bulunmuştur. CART algoritması ile prolin ve Brix seviyeleri sırasıyla %95 ve %100 doğrulukla sınıflandırdı (p< 0.001). Sonuç olarak, beyazımsı, mavimsi, siyahımsı ve yeşilimsi balların yüksek prolin ve düşük Brix içeriği nedeniyle daha kaliteli olduğu belirlenmiştir. Ancak renk özelliklerine dayalı balın kalitesini doğru bir şekilde değerlendirmek için daha fazla ve farklı orijinli bal örnekleri ile kapsamlı çalışmalara ihtiyaç vardır.

Anahtar Kelimeler

Tağşiş, Prolin, Brix, Balın kalitesi, CART algoritması

Etik Beyan

Çalışmada etik beyan gerektirecek şekilde herhangi bir hayvan veya insan kullanılmamıştır.

Destekleyen Kurum

Bu çalışma herhangi bir kurumdan destek alınmadan yürütülmüştür.

Teşekkür

Bal örneklerini sağlayan arı yetiştiricilerine teşekkür ederim.

Kaynakça

• Al-Farsi M, Al-Belushi S, Al-Amri A, Al-Hadhrami A, Al-Rusheidi M, Al-Alawi. Quality evaluation of Omani honey. Food Chem. 2018; 262: 162-167, https://doi.org/10.1016/j.foodchem.2018.04.104.
• Anonymous. Turkish Food Codex Honey Communiqué. https://www.resmigazete.gov.tr/eskiler/2020/04/20200422-13.htm, 2020, (Accessed: 10.04.2023).
• Anupama D, Bhat KK, Sapna VK. Sensory and physico-chemical properties of commercial samples of honey. Food Res. Int. 2003; 36(2): 183-191, https://doi.org/10.1016/S0963-9969(02)00135-7.
• Aykas DP. Determination of possible adulteration and quality assessment in commercial honey. Foods 2023; 12: Article 523, https://doi.org/10.3390/foods12030523.
• Aytekin İ, Eyduran E, Karadas K, Aksahan R, Keskin İ. Prediction of fattening final live weight from some body measurements and fattening period in young bulls of crossbred and exotic breeds using MARS data mining algorithm. Pak. J. Zool. 2018; 50(1): 189-195, http://dx.doi.org/10.17582/journal.pjz/2018.50.1.189.195.
• Bayır H. Determination of heavy metal level and some physicochemical properties in honey bee, honey and pollen produced in different locations of Konya province. Selcuk University. Graduate School of Natural and Applied Sciences, PhD Thesis, Konya, 2019, (Accessed: 01.02.2024), https://tez.yok.gov.tr/UlusalTezMerkezi/tezSorguSonucYeni.jsp.
• Bayram NE. Quality evaluation and pollen profile of honey samples from different locations. Prog. Nutr. 2019; 21(4): 928-934, https://doi.org/10.23751/pn.v21i4.8862.
• Becerril-Sánchez AL, Quintero-Salazar B, Dublán-García O, Escalona-Buendía HB. Phenolic compounds in honey and their relationship with antioxidant activity, botanical origin, and color. Antioxidants 2021; 10(11): Article 1700, https://doi.org/10.3390/antiox10111700.
• Bogdanov S. Harmonised methods of the international honey commission. Swiss Bee Research Centre FAM Liebefeld 2002; 5: 1-62.
• Boistean A, Chirsanova A, Capcanari T, Siminiuc R. Evaluation of the color as a characterization parameter of honey from Tunisia, Romania and Moldova. In: Biotehnologii moderne-soluții pentru provocările lumii contemporane; Chişinău; Moldova; 2021, p. 43-43. https://doi.org/10.52757/imb21.009.
• Breiman L, Friedman J, Olshen R, Stone C. Classification and regression trees. 1st ed. New York, NY, USA: Routledge; 1984, https://doi.org/10.1201/9781315139470.
• Cengiz MM, Tosun M, Topal M. Determination of the physicochemical properties and 13C/12C isotope ratios of some honeys from the northeast Anatolia region of Turkey. J. Food Compost. Anal. 2018; 69: 39-44, https://doi.org/10.1016/j.jfca.2018.02.007.
• Commission Internationale de l´Eclairage (CIE). Recommendations on uniform color spaces, color difference equations, psychometric color terms. Bureau Central de la CIE, Viena, 1978.
• Conti ME. Lazio region (central Italy) honeys: a survey of mineral content and typical quality parameters. Food Control 2000; 11(6): 459-463, https://doi.org/10.1016/S0956-7135(00)00011-6.
• Coşkun G, Şahin Ö, Delialioğlu RA, Altay Y, Aytekin İ. Diagnosis of lameness via data mining algorithm by using thermal camera and image processing method in Brown Swiss cows. Trop. Anim. Health Prod. 2023; 55: Article 50, https://doi.org/10.1007/s11250-023-03468-9.
• Dominguez MA, Centurión ME. Application of digital images to determine color in honey samples from Argentina. Microchem. J. 2015; 118: 110-114, https://doi.org/10.1016/j.microc.2014.08.002.
• Eker T, Bozdogan A, Ulukanlı Z. The physicochemical properties and antioxidant activities of honey from Kars (Turkey). In: III International Conference on Engineering and Natural Science; Budapest, Hungary; 2017, p. 1188-1192.
• Eyduran E. ehaGoF: Calculates goodness of fit statistics. R package version 0.1.1, 2020.
• Gaines-Day HR, Gratton C. Crop yield is correlated with honey bee hive density but not in high-woodland landscapes. Agric. Ecosyst. Environ. 2016; 218: 53-57, https://doi.org/10.1016/j.agee.2015.11.001.
• Geană EI, Ciucure CT, Costinel D, Ionete RE. Evaluation of honey in terms of quality and authenticity based on the general physicochemical pattern, major sugar composition and δ13C signature. Food Control 2020; 109: Article 106919, https://doi.org/10.1016/j.foodcont.2019.106919
• Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982; 143(1): 29-36, https://doi.org/10.1148/radiology.143.1.7063747.
• Haroun MI. Determination of phenolic and flavonoid profiles of some floral and honeydew honeys produced in Turkey. Ankara University, Graduate School of Natural and Applied Sciences, PhD Thesis, Ankara, 2006, (Accessed: 01.02.2024), https://tez.yok.gov.tr/UlusalTezMerkezi/tezSorguSonucYeni.jsp
• Hışıl Y, Börekçioğlu N. Composition of honey and adulterations. The J. Food 1986; 11(2): 79-82.
• IBM Corp. IBM SPSS statistics for windows, version 23.0. Armonk, NY: IBM Corp; 2015.
• Islam MK, Lawag IL, Green KJ, Sostaric T, Hammer KA et al. An investigation of the suitability of melissopalynology to authenticate Jarrah honey. Curr. Res. Food Sci. 2022; 5: 506-514, https://doi.org/10.1016/j.crfs.2022.02.014.
• Izquierdo M, Lastra-Mejías M, González-Flores E, Cancilla JC, Perez M, Torrecilla JS. Convolutional decoding of thermographic images to locate and quantify honey adulterations. Talanta 2020; 209: Article 120500, https://doi.org/10.1016/j.talanta.2019.120500.
• James OO, Mesubi MA, Usman LA, Yeye SO, Ajanaku KO, Ogunniran KO. Physical characterisation of some honey samples from North-Central Nigeria. Int. J. Phys. Sci. 2009; 4(9): 464-470, https://doi.org/10.5897/IJPS.9000439.
• Kanbur ED, Yuksek T, Atamov V, Ozcelik AE. A comparison of the physicochemical properties of chestnut and highland honey: The case of Senoz Valley in the Rize province of Turkey. Food Chem. 2021; 345: Article 128864, https://doi.org/10.1016/j.foodchem.2020,128864.
• Kapira K, Nkhata SG, Makolija N, Ayua EO, Aduol KO. Substituting natural honey for cane sugar (sucrose) retards microbial growth independent of water activity during storage of tomato jam. EJFOOD 2023; 5(1): 66-72, http://dx.doi.org/10.24018/ejfood.2023.5.1.609.
• Kayri M, Boysan M. Assesment of relation between cognitive vulnerability and depression's level by using classification and regression tree analysis. HU J. Educ. 2008; 34: 168-177.
• Kek SP, Chin NL, Yusof YA, Tan SW, Chua LS. Classification of entomological origin of honey based on its physicochemical and antioxidant properties. Int. J. Food Prop. 2017; 20(3): 2723-2738, https://doi.org/10.1080/10942912.2017.1359185.
• Khalafi R, Goli SAH, Behjatian M. Characterization and classification of several monofloral Iranian honeys based on physicochemical properties and antioxidant activity. Int. J. Food Prop. 2016; 19(5): 1065-1079, https://doi.org/10.1080/10942912.2015.1055360.
• Meda A, Lamien CE, Romito M, Millogo J, Nacoulma OG. Determination of the total phenolic, flavonoid and proline contents in Burkina Fasan honey, as well as their radical scavenging activity. Food Chem. 2005; 91(3): 571-577, https://doi.org/10.1016/j.foodchem.2004.10.006.
• Mehdi R, Zrira S, Vadalà R, Nava V, Condurso C, Cicero N et al. A Preliminary Investigation of Special Types of Honey Marketed in Morocco. JETA 2023; 1(1): 1-20, https://doi.org/10.3390/jeta1010001
• Mikail N, Keskin I. Subclinical mastitis prediction in dairy cattle by application of fuzzy logic. Pak. J. Agric. Sci. 2015; 52(4): 1101-1107.
• Nagai T, Kai N, Tanoue Y, Suzuki N. Chemical properties of commercially available honey species and the functional properties of caramelization and Maillard reaction products derived from these honey species. JFST 2018; 55: 586-597, https://doi.org/10.1007/s13197-017-2968-y.
• Polat G. Determination of the rheological, physicochemical characteristics and mineral contents of different floral honeys obtained from different locations. Selcuk University, Graduate School of Natural and Applied Sciences, Master Thesis, Konya, 2007, (Accessed: 01.02.2024), https://tez.yok.gov.tr/UlusalTezMerkezi/tezSorguSonucYeni.jsp.
• R Core Team. R: A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria, 2019, https://www.R-project.org/.
• Sancak K, Zan Sancak A, Aygören E. Dünya ve Türkiye’de arıcılık. Arıcılık Araştırma Dergisi 2013; 5: 7-13 (in Turkish).
• Shafiee S, Minaei S, Moghaddam-Charkari N, Ghasemi-Varnamkhasti M, Barzegar M. Potential application of machine vision to honey characterization. Trends Food Sci. Technol. 2013; 30(2): 174-177, https://doi.org/10.1016/j.tifs.2012.12.004.
• Tangirala S. Evaluating the impact of GINI index and information gain on classification using decision tree classifier algorithm. Int. J. Adv. Comput. Sci. Appl. 2020; 11(2): 612-619, https://doi.org/10.14569/IJACSA.2020.0110277.
• TSI. Turkish Statistical Institute. https://data.tuik.gov.tr/Kategori/GetKategori?p=tarim-111&dil=1, 2023, (Accessed: 15.02.2023).