A Big Data-Based Analysis: Color as a Predictor of Silage Quality

Yıl 2025, Cilt: 8 Sayı: 5, 725 - 735, 15.09.2025

Ömer Mermer , Ayşe Gül Filik , Gökhan Filik

https://doi.org/10.47115/bsagriculture.1721520

Öz

In this study, the relationship between color parameters and nutrient composition was examined using silage trial data conducted on 10 different plant species or mixtures by the Filik Research Lab. The dataset, comprising a total of 6,254 observations, was analyzed using big data analytics and statistical modeling techniques. Significant associations were identified between color parameters (L*, a*, b*, C*, h°, ΔE, saturation, and hue) and the physical, chemical, and sensory characteristics of the silage (crude protein, NDF, ADF, TDN, NFC). Initially, exploratory data analysis (EDA) and correlation matrices were employed to observe preliminary relationships between color and nutritional components. These were followed by classical and machine learning methods, such as multiple linear regression and Random Forest, to assess the strength and direction of these associations. The analysis revealed a* negative correlation (r = -0.45) between L* (lightness) and crude protein (CP, %), while a strong positive relationship (r = 0.52) was found between b* (yellowness) and NDF. Additionally, a meaningful correlation (r = 0.41) was observed between color saturation and TDN. Random Forest analysis confirmed these findings and highlighted L*, b*, and saturation as the most influential color parameters. These results suggest that color parameters are not only indicators of visual quality but also provide indirect insights into the chemical and nutritional content of silage. This study, conducted within the scope of big data analytics, demonstrates that color data can serve as a* rapid predictive tool, either as an alternative or complement to traditional chemical analyses. This presents a* significant contribution toward the development of digital quality control systems in agricultural production.

Anahtar Kelimeler

Silage quality , CIE Lab color space , Agricultural digitalization , Physical quality indicators , Machine learning , Agricultural decision support systems

Etik Beyan

Since no studies involving humans or animals were conducted, ethical committee approval was not required for this study.

Teşekkür

Some parts of this research were supported by individual projects funded by Kırşehir Ahi Evran University Scientific Research Commission. Additionally, certain experimental studies were carried out with the support of Filik Nanobioagrotech R&D and Technology Inc. The authors would like to thank Kırşehir Ahi Evran University, the Agricultural Faculty, the students of the Agricultural Biotechnology Department and also for the use of the University’s construction material laboratory for the execution of this research. The authors also acknowledge the administrative and technical support provided during the course of these studies.

Kaynakça

• Féret JB, de Boissieu F, Malenovský Z. 2020. PROSPECT-PRO for estimating content of nitrogen-containing leaf proteins and other carbon-based constituents. arXiv, 179. https://doi.org/10.48550/arXiv.2003.11961
• Filik G, Filik AG, Kezer G. 2022. As an alternative fermented feed for animal nutrition: Chia (Salvia). J Agric Sci, 26(1): 90-97. https://doi.org/10.15832/ankutbd.620982
• Filik G, Tekin OK, Filik AG, Çetinkaya O, Doğan Z, Çayan H, Şahin A. 2018. Yolk parameters in eggs of Atak-S parents. Int J Agric Nat Sci, pp:45-48.
• Fulgueira CL, Amigot SL, Gaggiotti M, Romero LA, Basílico JC. 2007. Forage quality: techniques for testing. Fresh Prod, 1(2): 121-131.
• Gao F, Fu L, Zhang X, Majeed Y, Li R, Karkee M, Zhang Q. 2020. Multi-class fruit-on-plant detection for apple in SNAP system using Faster R-CNN. Comput Electron Agric, 176: 105634. https://doi.org/10.1016/j.compag.2020.105634
• Geipel J, Bakken AK, Jørgensen M, Korsaeth A. 2021. Forage yield and quality estimation by means of UAV and hyperspectral imaging. Precis Agric, 22: 1437-1463. https://doi.org/10.1007/s11119-021-09790-2
• Goeser JP, Combs DK. 2009. An alternative method to assess 24-h ruminal in vitro neutral. Forage Grazelands, 7(1): 1-10. https://doi.org/10.1094/FG-2009-0223-01-RV
• Gonzalez RC, Woods RE. 2018 Digital Image Processing. 4th Edition, Pearson Education, New York, USA, pp: 1022.
• Hall MB, Herejk C. 2001. Differences in yields of microbial crude protein from in vitro fermentation of carbohydrates. J Dairy Sci, 84:(11), 2486-2493. https://doi.org/10.3168/jds.S0022-0302(01)74699-1 HunterLab. 1996. Hunter Lab color scale. Insight Color, 8(9): 1-15.
• İnce A, Vurarak Y. 2019. An Approach to Color Change and Quality Relation in Roughages. J Agric Sci, 25(1): 21-28. https://doi.org/10.15832/ankutbd.538982
• Kavlak AT, Pastell M, Uimari P. 2023. Disease detection in pigs based on feeding behaviour traits using machine learning. Biosyst Eng, 226: 132-143. https://doi.org/10.1016/j.biosystemseng.2023.01.004
• Kung L Jr, Shaver RD, Grant RJ, Schmidt RJ. 2018. Silage review: Interpretation of chemical, microbial, and organoleptic components of silages. J Dairy Sci, 101: 4020–4033. https://doi.org/10.3168/jds.2017-13909
• Mendoza F, Dejmek P, Aguilera JM. 2006. Calibrated color measurements of agricultural foods using image analysis. Postharv Biol Technol, 41(3): 285-295. https://doi.org/10.1016/j.postharvbio.2006.04.004
• Mertens DR. 1997. Creating a system for meeting the fiber requirements of dairy cows. J Dairy Sci, 80(7): 1463-1481. https://doi.org/10.3168/jds.S0022-0302(97)76075-2
• Tocco D, Carucci C, Monduzzi M, Salis A, Sanjust E. 2021. Recent developments in the delignification and exploitation of grass lignocellulosic biomass. ACS Sustainable Chem Eng, 9(6): 2412-2432. https://pubs.acs.org/doi/10.1021/acssuschemeng.0c07266
• Toruk F, Koç F, Gönülol E. 2010. Aerobik stabilite süresince paket silajlarinda renk değişimi. J Tekirdag Agric Fac, 7:(1): 23-30.
• Van Soest PJ, Robertson JB, Lewis BA. 1991. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci, 74(10): 3583-3597. https://doi.org/10.3168/jds.S0022-0302(91)78551-2
• Van Soest PJ. 1991. Nutritional ecology of the ruminant (2nd ed.). Cornell Univ Press, New York, USA, pp: 42-61.
• Wakholi C, Kim J, Nabwire S, Kwon KD, Mo C, Cho S, Cho BK. 2022. Deep learning feature extraction for image-based beef carcass yield estimation. Biosyst Eng, 218: 68-78. https://doi.org/10.1016/j.biosystemseng.2022.04.008
• Weiss WP, Conrad HR, St-Pierre NR. 1992. A theoretical-limit model for evaluating the nutritive value of forages. Anim Feed Sci Technol, 39(1-2): 95-110. https://doi.org/10.1016/0377-8401(92)90034-4