Ensemble of Deep Convolutional Neural Network Architecture for Classifying the Source and Quality of Teff Cereal

Authors: Belayneh Matebie, Michael Melese

Abstract:

The study focuses on addressing the challenges in classifying and ensuring the quality of Eragrostis Teff, a small and round grain that is the smallest cereal grain. Employing a traditional classification method is challenging because of its small size and the similarity of its environmental characteristics. To overcome this, the current study employs a machine learning approach to develop a source and quality classification system for Teff cereal. Data are collected from various production areas in the Amhara regions, considering two types of cereal (high and low quality) across eight classes. A total of 5,920 images are collected, with 740 images for each class. Image enhancement techniques, including scaling, data augmentation, histogram equalization, and noise removal, are applied to preprocess the data. Convolutional Neural Network (CNN) is then used to extract relevant features and reduce dimensionality. The dataset is split into 80% for training and 20% for testing. Different classifiers, including Fine-tunned Visual Geometry Group (FVGG16), Fine-tunned InceptionV3 (FINCV3), Quality and Source Classification of Teff Cereal (QSCTC), Ensemble Method for Quality and Source Classification of Teff Cereal (EMQSCTC), Support Vector Machine (SVM), and Random Forest (RF) are employed for classification, achieving accuracy rates ranging from 86.91% to 97.72%. The ensemble of FVGG16, FINCV3, and QSCTC using the Max-Voting approach outperforms individual algorithms.

Keywords: Teff, ensemble learning, Max-Voting, Convolutional Neural Network, Support Vector Machine, Random Forest.

References:

[1] E. Tadele and T. Hibistu, “Empirical review on the use dynamics and economics of teff in Ethiopia,” Agriculture and Food Security, vol. 10, no. 1. BioMed Central Ltd, Dec. 01, 2021. doi: 10.1186/s40066-021-00329-2.
[2] Z. Berhanu, M. Yifru, D. S. Geleta, and F. Liben, “Comparison of Teff (Eragrostis tef (Zucc.) Trotter) Varieties in Response to Blended NPSZnB Fertilizer On-Farm and on Research Station,” Advances in Agriculture, vol. 2023, 2023, doi: 10.1155/2023/2991419.
[3] “Agriculture and Food Security | Ethiopia | U.S. Agency for International Development.” Accessed: Mar. 13, 2022. Online. Available: https://www.usaid.gov/ethiopia/agriculture-and-food-security
[4] Y. A. Gebru, D. B. Sbhatu, and K. P. Kim, “Nutritional Composition and Health Benefits of Teff (Eragrostis tef (Zucc.) Trotter),” J Food Qual, vol. 2020, 2020, doi: 10.1155/2020/9595086.
[5] B. Ferede, F. Mekbib, K. Assefa, S. Chanyalew, E. Abraha, and Z. Tadele, “Evaluation of Tef (Eragrostis tef (Zucc.) Trotter) Somaclones for Drought Tolerance,” Advances in Crop Science and Technology, vol. 06, no. 04, pp. 4–11, 2018, doi: 10.4172/2329-8863.1000385.
[6] Asmiro. A. Fikadu, Tsega. D. Wedu, and Endalew. A. Derseh, “Review on Economics of Teff in Ethiopia,” Open Acc Biostatistics & Bioinformatics, vol. 2, no. 3, pp. 1–8, 2019, doi: 10.31031/OABB.2018.02.000539.
[7] A. Abewa et al., “Teff grain physical and chemical quality responses to soil physicochemical properties and the environment,” Agronomy, vol. 9, no. 6, 2019, doi: 10.3390/agronomy9060283.
[8] B. A. Gashaw and S. M. Kibret, “The Role of Ethiopian Commodity Exchange (ECX) in Crop Value Chain Development in Ethiopia,” no. October, 2018, doi: 10.11648/j.jiber.20180706.12.
[9] B. G. Asefa, F. Tsige, M. Mehdi, T. Kore, and A. Lakew, “Rapid classification of tef (Eragrostis tef (Zucc.) Trotter) grain varieties using digital images in combination with multivariate technique,” Smart Agricultural Technology, vol. 3, Feb. 2023, doi: 10.1016/j.atech.2022.100097.
[10] M. Punn and N. Bhalla, “Classification of Wheat Grains Using Machine Algorithms,” vol. 2, no. 8, pp. 363–366, 2013.
[11] J. R. Arunkumar and T. Mengist, “Developing Ethiopian Yirgacheffe Coffee Grading Model using a Deep Learning Classifier,” International Journal of Innovative Technology and Exploring Engineering, vol. 9, no. 4, pp. 3303–3309, 2020, doi: 10.35940/ijitee.d1823.029420.
[12] A. L. Vvcxs, “College of Natural Sciences Development of Automatic Maize Quality Assessment System Using Image Processing Techniques,” no. November, 2015.
[13] C. Sewwandi Silva and U. Sonnadara, “Classification of Rice Grains Using Neural Networks,” Proceedings of Technical Sessions, vol. 29, no. June, pp. 9–14, 2013.
[14] K. Herath, “Rice Grains Classification Using Image Processing Technics,” no. January, 2017.
[15] S. A. Villar, S. Torcida, and G. G. Acosta, “Median Filtering: A New Insight,” J Math Imaging Vis, vol. 58, no. 1, pp. 130–146, May 2017, doi: 10.1007/s10851-016-0694-0.
[16] “A Comprehensive Guide on Deep Learning Optimizers.” Accessed: Jun. 01, 2022. Online. Available: https://www.analyticsvidhya.com/blog/2021/10/a-comprehensive-guide-on-deep-learning-optimizers/
[17] A. R. Baleker and A. Ababa, “Raw Quality Value Classification of Ethiopian Coffee Using Image Processing Techniques: In the case of Wollega region,” 2011.
[18] M. F. Abeza, “Development of an Automated Grading System of White Pea Beans Using Image Processing Techniques Convergence,” no. August, 2019.