Derin öğrenmeye dayalı görünür yakın kızılötesi kamera kullanılarak buğday sınıflandırması

Öz

Bu makale, derin öğrenme metodolojisine dayalı hiperspektral buğday verilerinin sınıflandırılması için akıllı bir makine öğrenme sistemi sunmaktadır. Bu amaçla, hiperspektral buğday örneklerinin sınıflandırılması için AlexNet ve VGG16 modellerinin performansları araştırılmıştır. Bu çalışmada, buğday çekirdeklerinin türlerini tahmin etmek için Destek Vektör Makinesi (DVM) ve Softmax sınıflandırıcıları kullanılmıştır. Sistem performansını değerlendirmek için, Görünür Yakın Kızılötesi Görüntüleme (VNIR) kullanılarak 50 buğday türüne ait tür başına 220 görüntü toplamda 11000 örnek içeren yeni bir hiperspektral buğday test veri kümesi oluşturulmuştur. Yeni oluşturulan test veri seti üzerinde yapılan deneylerde, AlexNet ve VGG16 için tamamen bağlı katman (FC6 ve FC7) özellikleri kullanılması durumunda doğrusal DVM sınıflandırıcısı tarafından belirlenen yaklaşık %96.00 ve % 99.00'lık genel doğruluk oranları elde edilmiştir. Softmax sınıflandırıcı ile numunelerin sırasıyla %92 ve %70'i, eğitimli VGG16 ve AlexNet modellerine göre doğru bir şekilde ayırt edilebilmiştir. Elde edilen üstün sonuçlar, derin bir Evrişimsel Sinir Ağları (ESA) mimarisi kullanmanın, buğday türlerinin doğru bir şekilde ayırt edilmesi yoluyla daha verimli olduğunu göstermektedir. Önerilen derin öğrenme temelli sınıflandırma sistemi, gıdalarda kalite analizi, sınıflandırma ve hastalık tespiti için yüksek doğrulukta sonuçlar vaat etmektedir.

Anahtar Kelimeler

• Besin mühendisliği
• Evrişimsel sinir ağları
• Sınıflandırma
• Görünür yakin kızılötesi
• Buğday

Wheat kernels classification using visible-near infrared camera based on deep learning

Abstract

This paper presents a smart machine learning system for classification of hyperspectral wheat data based on deep learning methodology. For this purpose, the performances of AlexNet and VGG16 models were investigated for the classification of hyperspectral wheat samples. In this study, the Support Vector Machine (SVM) and Softmax classifiers were carried out to predict labels of wheat kernels. In order to evaluate the system performance, a new hyperspectral wheat test dataset was constructed using Visible-Near Infrared images (VNIR) including 50 wheat species with 220 images per specimen, as 11000 samples in total. With experiments applied on newly created test dataset, overall approximated accuracy rates of 96.00% and 99.00% determined by linear SVM classifier, in case of fully connected layer (FC6 and FC7) features for AlexNet and VGG16, respectively. From the Softmax predictions, the 92% and 70% of samples were correctly discriminated based on trained VGG16 and AlexNet models, respectively. The obtained superior results show that using a deep Convolutional Neural Networks (CNN) architecture is more efficient by the means of accurate discrimination of wheat species. The proposed deep learning based categorization system promises high accuracy results for the quality analysis, classification and disease detection in food.

Keywords

• Food engineering
• Convolutional neural network
• Classification
• Visible-Near infrared
• Wheat

References

1. [1] Agricultural Research Institute. "Directorate of Trakya Agricultural Research Institute". https://Arastirma.Tarimorman.Gov.Tr/Ttae/Sayfalar/De tay.Aspx?Sayfaid=47 (26.03.2021).
2. [2] Charytanowicz M, Kulczycki P, Kowalski PA, Łukasik S, Czabak-Garbacz R. "An evaluation of utilizing geometric features for wheat grain classification using x-ray images". Computers and Electronics in Agriculture, 144, 260-268, 2018.
3. [3] Vermeulen P, Michele S, Juan PFA, Vincent B. "Discrimination between durum and common wheat kernels using near infrared hyperspectral imaging". Journal of Cereal Science, 84, 74-82 2018.
4. [4] Miralbés C. "Discrimination of european wheat varieties using near infrared reflectance spectroscopy". Food Chemistry, 106(1), 386-389, 2008.
5. [5] Mahesh S, Manickavasagan A, Jayas D, Paliwal J, White N. "Feasibility of near-infrared hyperspectral imaging to differentiate canadian wheat classes". Biosystems Engineering, 101(1), 50-57, 2008.
6. [6] Singh C, Jayas D, Paliwal J, White N. "Detection of insectdamaged wheat kernels using near-infrared hyperspectral imaging". Journal of Stored Products Research, 45(3), 151-158, 2009.
7. [7] Mutlu AC, Boyaci IH, Genis HE, Ozturk R, Basaran N, Sanal T, Evlice AK. "Prediction of wheat quality parameters using near-infrared spectroscopy and artificial neural networks". European Food Research and Technology, 233(2), 267-274, 2011.
8. [8] Guevara FH, Gomez JG. "A machine vision system for classification of wheat and barley grain kernels". Spanish Journal of Agricultural Research, 9(3), 672-680, 2011.

9. [9] Soto MC, Gaitán AJ, Domínguez J. "Application of near infrared spectroscopy technology for the detection of fungicide treatment on durum wheat samples". Talanta, 97, 298-302, 2012.
10. [10] Serranti S, Cesare D, Bonifazi G. "The development of a hyperspectral imaging method for the detection of fusarium-damaged, yellow berry and vitreous italian durum wheat kernels". Biosystems Engineering, 115(1), 20-30, 2013.
11. [11] González MIM, Moncada GW, González CP, San NZM, López FG, Ortega IL, Hernández JMH. "Chilean flour and wheat grain: tracing their origin using near infrared spectroscopy and chemometrics". Food Chemistry, 145, 802-806, 2014.
12. [12] Jaillais B, Roumet P, Pinson-Gadais L, Bertrand D. "Detection of fusarium head blight contamination in wheat kernels by multivariate imaging". Food Control, 54, 250-258, 2015.
13. [13] Ziegler JU, Leitenberger M, Longin CFH, Würschum T, Carle R, Schweiggert RM. "Near-Infrared reflectance spectroscopy for the rapid discrimination of kernels and flours of different wheat species". Journal of Food Composition and Analysis, 51, 30-36, 2016.
14. [14] Pound MP, Atkinson JA, Townsend AJ, Wilson MH, Griffiths M, Jackson AS, Bulat A, Tzimiropoulos G, Wells DM, Murchie EH, Pridmore TP. "Deep machine learning provides state-of-the-art performance in image-based plant phenotyping". Gigascience, 2018. https://doi.org/10.1093/gigascience/giy042.
15. [15] Chatnuntawech I, Tantisantisom K, Khanchaitit P, Boonkoom T, Bilgic B, Chuangsuwanich E. "Rice classification using spatio-spectral deep convolutional neural network". ArXiv, 2018. https://arxiv.org/abs/1805.11491.
16. [16] Simonyan K, Zisserman A. "Very deep convolutional networks for large-scale image recognition". ArXiv, 2014. https://arxiv.org/abs/1409.1556.
17. [17] He K, Zhang X, Ren S, Sun J. "Deep residual learning for image recognition". IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, USA, 12 December 2016.
18. [18] Qiu Z, Chen J, Zhao Y, Zhu S, He Y, Zhang C. "Variety identification of single rice seed using hyperspectral imaging combined with convolutional neural network". Journal of Applied Sciences, 8(2), 1-12, 2018.
19. [19] Ni C, Wang D, Vinson R, Holmes M, Tao Y. "Automatic inspection machine for maize kernels based on deep convolutional neural networks". Biosystems Engineering, 178, 131-144, 2019.
20. [20] Özkan K, Isik S, Yavuz B, Demirez DZ. "Shallow and deep convolutional neural network models for classification of VNIR wheat samples". Fifth International Conference on Engineering and Natural Sciences, Prague, Czech Republic, 12-16 June 2019.
21. [21] Özkan K, Isik S, and Yavuz BT. "Identification of wheat kernels by fusion of RGB, SWIR, and VNIR samples". Journal of the Science of Food Agriculture, 99(11), 4977-4984, 2019.
22. [22] Ahonen T, Hadid A, Pietikainen M. "Face description with local binary patterns: application to face recognition". IEEE Transactions on Pattern Analysis Machine Intelligence, 28(12), 2037-2041, 2006.
23. [23] Dalal N, Triggs B. "Histograms of oriented gradients for human detection". Fifth IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, USA, 20-25 June 2005.
24. [24] Lowe DG. "Distinctive ımage features from scale-invariant keypoints". International Journal of Computer Vision, 60(2), 91-110, 2004.
25. [25] Krizhevsky A, Sutskever I, Hinton GE. "Imagenet classification with deep convolutional neural networks". Advances in Neural Information Processing Systems, 25, 1097-1105, 2012.
26. [26] Chang CC, Lin CJ. "LIBSVM: a library for support vector machines". Journal of ACM Transactions on Intelligent Systems Technology, 2(3), 1-27, 2011.
27. [27] Olgun M, Onarcan AO, Özkan K, Isik S, Sezer O, Özgişi K, Ayter NG, Başçiftçi ZB, Ardiç M, Koyuncu O. "Wheat grain classification by using dense SIFT features with SVM classifier". Computers and Electronics in Agriculture, 122, 185-190, 2016.
28. [28] Williams PJ, Kucheryavskiy S. "Classification of maize kernels using NIR hyperspectral imaging". Food Chemistry, 209, 131-138, 2016.
29. [29] Fan Y, Ma S, Wu T. "Individual wheat kernels vigor assessment based on NIR spectroscopy coupled with machine learning methodologies". Infrared Physics Technology, 2020. doi:10.1016/j.infrared.2020.103213.