Classification of Some Barley Cultivars with Deep Convolutional Neural Networks

Year 2023, Volume: 29 Issue: 1, 262 - 271, 31.01.2023

Fatih Bayram , Mustafa Yıldız

https://doi.org/10.15832/ankutbd.815230

Cited By: 1

Abstract

The homogeneity of the seeds is an important factor in terms of processing, transportation, storage, and product quality of agricultural products. It is possible to classify the grain polymorphism of barley cultivars, which are economically important among cereal crops, in a short time with computer vision methods with high accuracy rate and almost zero cost. In this research, a novel image database consisting of 2800 images were created to classify 14 barley cultivars. Six different deep convolutional neural network models were designed based on a transfer learning method with pretrained DenseNet-121, DenseNet-169, DenseNet-201, InceptionResNetV2, MobileNetV2 and Xception networks. The models were trained and evaluated with test-time augmentation method, the best performance was obtained from DenseNet-169 model with average 96.07% recall, 96.29% precision, 96.07% F1-score, and 96.07% accuracy on a test set independent of the training set. The results showed that the transfer learning method performed using additional layers such as dropout and data augmentation with sufficient data samples in these images with high similarities prevented overfitting by increasing the model performance. As a result, it can be suggested that the provided web tool based on the transfer model has an encouraging performance in identifying seeds
with a high number of cultivars such as barley. 

Keywords

Computer vision, Automatic seed recognition, Transfer learning, Barley classification

References

• Bothmer R (1992). The wild species of Hordeum: relationships and potential use for improvement of cultivated barley. In: P R Shewry (Eds.), Barley: Genetics, biochemistry, molecular biology and biotechnology, CAB Int, Wallingford, pp. 3-18
• Choudhary R, Paliwal J & Jayas D (2008). Classification of cereal grains using wavelet, morphological, colour, and textural features of non-touching kernel images. Biosystems Engineering 99(3): 330–337
• Dolata P & Reiner J (2018). Barley variety recognition with viewpoint-aware double-stream convolutional neural networks. In: 2018 Federated Conference on Computer Science and Information Systems (FedCSIS), 9-12 September, Poznan, pp. 101–105
• El Rabey H, Al-Malki A, Abulnaja K, Ebrahim M, Kumosani, T & Khan J (2014). Phylogeny of ten species of the genus Hordeum L. as revealed by AFLP markers and seed storage protein electrophoresis. Molecular Biology Reports 41(1): 365–372 Hailu B & Meshesha M (2016). Applying image processing for malt-barley seed identification. In: Ethiopian the 9th ICT Annual Conference, 2-6 June, Addis Ababa, pp. 10-16
• Huang G, Liu Z, Van Der Maaten L & Weinberger K Q (2017). Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 21-26 July, Honolulu, pp. 2261–2269
• Kozłowski M, Górecki P & Szczypiński P (2019). Varietal classification of barley by convolutional neural networks. Biosystems Engineering 184:155–165
• Kün E (1988). Serin iklim tahılları. Ankara Üniversitesi Ziraat Fakültesi Yayınları, 1032, Ankara
• LeCun Y, Bottou L, Bengio Y & Haffner P (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE 86(11): 2278–2324
• Majumdar S & Jayas D (2000a). Classification of cereal grains using machine vision: I. Morphology models. Transactions of the ASAE 43(6): 1669
• Majumdar S & Jayas D (2000b). Classification of cereal grains using machine vision: II. Color models. Transactions of the ASAE 43(6): 1677
• Majumdar S & Jayas D (2000c). Classification of cereal grains using machine vision: III. Texture models. Transactions of the ASAE 43(6): 1681 Majumdar S & Jayas D (2000d). Classification of cereal grains using machine vision: IV. Combined morphology, color, and texture models Transactions of the ASAE 43(6): 1689
• McCulloch W S & Pitts W (1943). A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics 5(4): 115–133
• Mebatsion H K, Paliwal J & Jayas D S (2013). Automatic classification of non-touching cereal grains in digital images using limited morphological and color features. Computers and Electronics in Agriculture 90: 99–105
• Neuman M, Sapirstein H, Shwedyk E & Bushuk W (1987). Discrimination of wheat class and variety by digital image analysis of whole grain samples. Journal of Cereal Science 6(2): 125–132
• Paliwal J, Visen N & Jayas D (2001). Evaluation of neural network architectures for cereal grain classification using morphological features. Journal of Agricultural Engineering Research 79(4): 361–370
• Szczypiński P M, Klepaczko A & Zapotoczny P (2015). Identifying barley varieties by computer vision. Computers and Electronics in Agriculture 110: 1–8
• Zapotoczny P, Zielinska M & Nita Z (2008). Application of image analysis for the varietal classification of barley: Morphological features. Journal of Cereal Science 48(1): 104-110