Tahribatsız Fiziksel ve Mekanik Ölçümler Kullanılarak Ceviz İç Ağırlığının Karşılaştırmalı Modellenmesi

Abstract

Bir ürün tüketiciye ulaşana kadar birçok aşamadan geçer. Ürünün yetiştirilmesi hasatla tamamlanırken, tüketiciye en iyi şekilde ulaşması hasat sonrasındaki sürece bağlıdır. Hasat sonrası süreç, taşıma, depolama, işleme, paketleme vb. işlemler birçok faktörün etkisi altında gerçekleşir. Bu faktörlerin etkisini azaltmak veya bu faktörlerin etkisini azaltmak için önlemler almak amacıyla ürünün özelliklerini bilmek çok önemlidir. Uzun yıllardır tüketiciler, meyve ve kabuklu yemişlerin dış özelliklerine büyük önem vermektedir. Ancak son yıllarda tüketicilerin meyve ve kabuklu yemişlerin iç kalite özelliklerine olan ilgisi de artmıştır. Üreticiler ve pazarlamacılar için en büyük zorluk, ürünlerin iç kalite özelliklerini tahribatsız bir şekilde belirleyebilmektir. Bu nedenle, günümüzde tahribatsız değerlendirme teknikleri büyük önem kazanmış ve ürünlerin iç ve dış kalite göstergelerinin tahmin edilmesine olanak sağlamıştır. Bu çalışmanın amacı, ceviz iç ağırlığını tahribatsız bir şekilde modellemektir. Bu amaç doğrultusunda çoklu doğrusal regresyon (MLR), temel bileşen analizi (PCA), yapay sinir ağı (ANN), temel bileşen analizi + yapay sinir ağı (PCA + ANN), derin öğrenme sinir ağı (DLNN) ve destek vektör makineleri (SVM) gibi farklı yöntemler test edilmiştir. Bu modeller için giriş değişkenleri; çeşit, ceviz uzunluğu (mm), ceviz genişliği (mm), kalınlık (mm), ceviz ağırlığı (g), geometrik ortalama çap (mm), küresellik (%) ve kırılma kuvveti (N) olarak belirlenmiştir. Araştırma sonuçlarına göre, ceviz iç ağırlığını tahmin etmede en yüksek model doğruluğu %98.14 ile SVM yöntemiyle elde edilmiştir. ANN ve DLNN yöntemlerinde model doğruluk oranları sırasıyla %89.78 ve %93.49 olarak belirlenmiştir. Hibrit model (PCA+ANN) kullanıldığında ise doğruluk oranı %58.8’e düşmüştür. Ayrıca, MLR yönteminde başarı oranı %76.43 olarak tespit edilmiştir. Genel bulgular, diğer makine öğrenmesi teknikleri ile karşılaştırıldığında SVM yaklaşımının ceviz iç ağırlığının tahmini açısından daha doğru sonuçlar ürettiğini ortaya koymaktadır.

Keywords

• Ceviz içi
• Modelleme
• Çoklu doğrusal regresyon (MLR)
• Yapay sinir ağı (ANN)
• Derin öğrenme sinir ağı (DLNN)

Comparative Modelling of Walnut Kernel Weight Using Non-Destructive Physical and Mechanical Measurements

Abstract

A product undergoes several stages before reaching the consumer. Although cultivation is completed at harvest, ensuring that the product reaches consumers under optimal conditions is equally important. Postharvest processes, including transportation, storage, processing, and packaging, are influenced by numerous factors. Understanding the intrinsic characteristics of agricultural products is therefore essential for minimizing the adverse effects of these factors and for implementing appropriate management strategies. For many years, consumer preferences have primarily focused on the external appearance of fruits and nuts. In recent years, however, increasing attention has been directed toward internal quality attributes. Consequently, a major challenge for growers and marketers is the reliable assessment of internal quality characteristics without causing damage to the product. In this context, non-destructive techniques have gained considerable importance, as they enable the evaluation and prediction of internal quality parameters based on measurable external properties. The present study aimed to model walnut kernel weight using non-destructive approaches. To this end, several modelling techniques were evaluated, including multiple linear regression (MLR), principal component analysis (PCA), artificial neural networks (ANN), PCA combined with ANN (PCA+ANN), deep learning neural networks (DLNN), and support vector machines (SVM). The input variables included variety, walnut length (mm), width (mm), thickness (mm), walnut weight (g), geometric mean diameter (mm), sphericity (%), and cracking force (N). The results demonstrated that the SVM model achieved the highest predictive performance for walnut kernel weight estimation, with an accuracy of 98.14%. The ANN and DLNN models yielded accuracy values of 89.78% and 93.49%, respectively. In contrast, the hybrid PCA+ANN model exhibited a substantial decline in performance, achieving an accuracy of 58.8%. The MLR model produced an accuracy of 76.43%. Overall, the findings indicate that, among the evaluated machine learning techniques, the SVM approach provides the most accurate and robust predictions for non-destructive estimation of walnut kernel weight.

Keywords

• Kernel
• Modelling
• Multiple linear regression (MLR)
• Artificial neural network (ANN)
• Deep learning neural network (DLNN)

References

1. Abdel-Sattar, M., Aboukarima, A. M. and Alnahdi, B. M. (2021). Application of artificial neural network and support vector regression in predicting mass of ber fruits (Ziziphus mauritiana Lamk.) based on fruit axial dimensions. PLOS ONE, 16: e0245228. https://doi.org/10.1371/journal.pone.0245228
2. Ahmadi, H. and Rodehutscord, M. (2017). Application of artificial neural network and support vector machines in predicting metabolizable energy in compound feeds for pigs. Frontiers in Nutrition, 4: 27. https://doi.org/10.3389/fnut.2017.00027
3. An, M., Cao, C., Wu, Z. and Luo, K. (2022). Detection method for walnut shell-kernel separation accuracy based on near-infrared spectroscopy. Sensors, 22: 8301. https://doi.org/10.3390/s22218301
4. Anagnostis, A., Tagarakis, A. C., Asiminari, G., Papageorgiou, E., Kateris, D., Moshou, D. and Bochtis, D. (2021). A deep learning approach for anthracnose infected trees classification in walnut orchards. Computers and Electronics in Agriculture, 182: 105998. https://doi.org/10.1016/j.compag.2021.105998
5. Anuse, A. and Vyas, V. (2016). A novel training algorithm for convolutional neural network. Complex and Intelligent Systems, 2: 221-234. https://doi.org/10.1007/s40747-016-0024-6
6. Arendse, E., Fawole, O. A., Magwaza, L. S. and Opara, U. L. (2018). Non-destructive prediction of internal and external quality attributes of fruit with thick rind: A review. Journal of Food Engineering, 217: 11-23. https://doi.org/10.1016/j.jfoodeng.2017.08.009
7. Aydin, C. (2003). Physical properties of almond nut and kernel. Journal of Food Engineering, 60(3): 315-320. https://doi.org/10.1016/S0260-8774(03)00053-0
8. Çelekli, A., Birecikligil, S. S., Geyik, F. and Bozkurt, H. (2012). Prediction of removal efficiency of Lanaset Red G on walnut husk using artificial neural network model. Bioresource Technology, 103: 64-70. https://doi.org/10.1016/j.biortech.2011.09.106

9. Ceylan, İ. and Aktaş, M. (2008). Modelling of a hazelnut dryer assisted heat pump by using artificial neural networks. Applied Energy, 85(9): 841-854. https://doi.org/10.1016/j.apenergy.2007.10.013
10. Ciftci, K. and Gökce, O. (2006). A research on the socio-economical aspects and problems of walnut production in İzmir and Manisa. Yuzuncu Yıl University Journal of Agricultural Sciences, 16(1): 7-17.
11. Colaric, M., Veberic, R., Solar, A., Hudina, M. and Stampar, F. (2005). Phenolic acids, syringaldehyde, and juglone in fruits of different cultivars of Juglans regia L. Journal of Agricultural and Food Chemistry, 53: 6390-6396. https://doi.org/10.1021/jf050721n
12. Daley, O., Isaac, W. A., John, A. and Kahramanoğlu, İ. (2024). Preharvest factors, maturity indices and postharvest physiology of fruits and vegetables. In Postharvest Physiology and Handling of Horticultural Crops (pp. 3-35). CRC Press.
13. Desai, K. M., Survase, S. A., Saudagar, P. S., Lele, S. S. and Singhal, R. S. (2008). Comparison of artificial neural network (ANN) and response surface methodology (RSM) in fermentation media optimization: case study of fermentative production of scleroglucan. Biochemical Engineering Journal, 41: 266-273.
14. Ercisli, S., Kara, M., Ozturk, I., Sayinci, B. and Kalkan, F. (2011). Comparison of some physico-mechanical nut and kernel properties of two walnut (Juglans regia L.) cultivars. Notulae Botanicae Horti Agrobotanici Cluj-Napoca, 39(2): 227-231. https://doi.org/10.15835/nbha3926045
15. Gardner, M. W. and Dorling, S. R. (1998). Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric Environment, 32(14-15): 2627-2636. https://doi.org/10.1016/S1352-2310(97)00447-0
16. Grace, M. H., Warlick, C. W., Neff, S. A. and Lila, M. A. (2014). Efficient preparative isolation and identification of walnut bioactive components using high-speed counter-current chromatography and LC-ESI-IT-TOF-MS. Food Chemistry, 158, 229-238. DOI: 10.1016/j.foodchem.2014.02.117
17. Huang, K. Y. (2012). Detection and classification of areca nuts with machine vision. Computers and Mathematics with Applications, 64(5): 739-746. https://doi.org/10.1016/j.camwa.2011.11.041
18. Jin, F., Qin, L., Rao, X. and Tao, Y. (2007). Walnut shell and meat classification using texture analysis and SVMs. In Optics for Natural Resources, Agriculture, and Foods. 2: 149-160. https://doi.org/10.1117/12.730907
19. Kahramanoğlu, İ., Ülker, E. D. and Ülker, S. (2021). Integrating support vector machine (SVM) technique and contact imaging for fast estimation of the leaf chlorophyll contents of strawberry plants. International Journal of Engineering Trends and Technology, 69(3): 23-28. https://doi.org/10.14445/22315381/IJETT-V69I3P205
20. Kavuncuoglu, H., Kavuncuoglu, E., Karatas, S. M., Benli, B., Sagdic, O. and Yalcin, H. (2018). Prediction of the antimicrobial activity of walnut (Juglans regia L.) kernel aqueous extracts using artificial neural network and multiple linear regression. Journal of Microbiological Methods, 148: 78-86. DOI: 10.1016/j.mimet.2018.04.003 Khalesi, S., Mahmoudi, A., Hosainpour, A. and Alipour, A. (2012). Detection of walnut varieties using impact acoustics and artificial neural networks (ANNs). Modern Applied Science, 6: 43. https://doi.org/10.5539/mas.v6n1p43
21. Khataee, A. R., Dehghan, G., Zarei, M., Ebadi, E. and Pourhassan, M. (2011). Neural network modelling of biotreatment of triphenylmethane dye solution by a green macroalgae. Chemical Engineering Research and Design, 89: 172-178. https://doi.org/10.1016/j.cherd.2010.05.009
22. Khezri, V., Yasari, E., Panahi, M. and Khosravi, A. (2020). Hybrid artificial neural network–genetic algorithm-based technique to optimize a steady-state gas-to-liquids plant. Industrial and Engineering Chemistry Research, 59: 8674-8687. https://hdl.handle.net/10536/DRO/DU:30139017
23. Khorami, S. S., Arzani, K., Karimzadeh, G., Shojaeiyan, A. and Ligterink, W. (2018). Genome size: A novel predictor of nut weight and nut size of walnut trees. HortScience, 53(3): 275-282. https://doi.org/10.21273/HORTSCI12725-17
24. Magwaza, L. S. and Opara, U. L. (2015). Analytical methods for determination of sugars and sweetness of horticultural products—A review. Scientia Horticulturae, 184: 179-192. https://doi.org/10.1016/j.scienta.2015.01.001
25. Naroui Rad, M. R., Ghalandarzehi, A. and Koohpaygani, J. A. (2017). Predicting eggplant individual fruit weight using an artificial neural network. International Journal of Vegetable Science, 23(4): 331-339. https://doi.org/10.1080/19315260.2017.1290001
26. Nicolai, B. M., Beullens, K., Bobelyn, E., Peirs, A., Saeys, W., Theron, K. I. and Lammertyn, J. (2007). Non-destructive measurement of fruit and vegetable quality by means of NIR spectroscopy: A review. Postharvest Biology and Technology, 46: 99-118. https://doi.org/10.1016/j.postharvbio.2007.06.024
27. Ojaghloo, B., Vahdati, K., Amiri, R. and Hassani, D. (2014). A statistical model to estimate potential yields in Persian walnut (Juglans regia L.). Journal of Biodiversity and Environmental Sciences, 4(5): 1-9.
28. Omid, M., Baharlooei, A. and Ahmadi H. (2009). Modeling drying kinetics of pistachio nuts with multilayer feed-forward neural network. Drying Technology, 27: 1069-1077. https://doi.org/10.1080/07373930903218602
29. Parveen, N., Zaidi, S. and Danish, M. (2017). Development of SVR-based model and comparative analysis with MLR and ANN models for predicting the sorption capacity of Cr (VI). Process Safety and Environmental Protection, 107: 428-437. https://doi.org/10.1016/j.psep.2017.03.007
30. Rong, D., Xie, L. and Ying, Y. (2019). Computer vision detection of foreign objects in walnuts using deep learning. Computers and Electronics in Agriculture, 162: 1001-1010. https://doi.org/10.1016/j.compag.2019.05.019
31. Solmaz, Y. and Adiloğlu, A. (2017). Determination of nutritional status of walnut orchards by leaf analysis in Tekirdag Region. Journal of Tekirdag Agricultural Faculty, 14(1): 88-92.
32. Suykens, J. A. and Vandewalle, J. (1999). Least squares support vector machine classifiers. Neural Processing Letters, 9: 293-300.
33. Taşan, M. and İmer, Y. (2018). Determination of some micro and macronutrient elements in various cold press vegetable oils. Journal of Tekirdag Agricultural Faculty, 15(1): 14-25. (In Turkish)
34. Teimouri, N., Omid, M., Mollazade, K. and Rajabipour, A. (2014). A novel artificial neural network assisted segmentation algorithm for discriminating almond nut and shell from background and shadow. Computers and Electronics in Agriculture, 105: 34-43. https://doi.org/10.1016/j.compag.2014.04.008
35. Vapnik, V. and Izmailov, R. (2021). Reinforced SVM method and memorization mechanisms. Pattern Recognition, 119: 108018. https://doi.org/10.1016/j.patcog.2021.108018
36. Yang, W. Y., Cao, W., Kim, J., Park, K. W., Park, H. H., Joung, J., Ro, J. S., Lee, H. L., Hong, C. H. and Im, T. (2020). Applied numerical methods using MATLAB. John Wiley and Sons. https://doi.org/10.1002/9781119626879
37. Zhai, Z., Jin, Z., Li, J., Zhang, M. and Zhang, R. (2020). Machine learning for detection of walnuts with shrivelled kernels by fusing weight and image information. Journal of Food Process Engineering, 43: e13562. https://doi.org/10.1111/jfpe.13562