Dalgacık Görüntü Saçılımı ve DenseNet Temelli Fıstık Tanılaması

Öz

Günümüzde tarım sektöründe ürünlerin ekonomik değerlerinin ve endüstriyel süreçlerin verimliliğinin artırılması ve zirai ürünlerin birbirinden ayırt edilmesi için bilgisayar temelli sistemler önem kazanmaktadır. Ülkemizde yetiştirilen Kırmızı ve Siirt tipi fıstık çeşitleri fiyat, besin değeri, şekil, lezzet gibi birçok yönden birbirinden farklıdır. Bu çalışmada, ülkemizde yetişen Kırmızı ve Siirt fıstık çeşitlerini ayırt etmek için dalgacık görüntü saçılımı ve DarkNet53 evrişimsel sinir ağına (ESA) dayanan bir sınıflandırma modeli geliştirilmiştir. Çalışma kapsamında 1232 Kırmızı ve 916 Siirt çeşidi olmak üzere toplamda 2148 fıstık çeşitlerinin görüntüleriyle çalışma gerçekleşmiştir. Bu görüntüleri sınıflandırmak için dalgacık görüntü saçılımı ve DarkNet53 evrişimsel sinir ağı mimarisi ile görüntülere ait özellikler elde edilmiştir ve ardından bu özellikler Destek Vektör Makinaları (DVM) ile sınıflandırılmıştır. Dalgacık görüntü saçılımı ve DarkNet53 ESA mimarisi kullanılarak görüntülere ait oluşturulan özellik setinin DVM ile sınıflandırma sonucu %97.98 doğruluk elde edilmiştir.

Anahtar Kelimeler

• Kırmızı fıstık
• Siirt fıstığı
• Dalgacık görüntü saçılımı
• DarkNet53
• Destek vektör makinaları.

Image Wavelet Scattering and Densenet Based Pistachio Identification

Abstract

Today, computer-based systems are gaining importance in the agricultural sector in order to increase the economic value of products, industrial processing efficiency, and recognition of agricultural products. Pistacia vera (Kırmızı and Siirt pistachio) varieties grown in Turkey differ from each other in many ways such as price, nutritional value, shape and flavor. In this study, a classification model based on wavelet image scattering and DarkNet53 convolutional neural network (ESA) was developed to distinguish the Red and Siirt pistachio cultivars grown in our country. Within the scope of the study, the study was carried out with images of a total of 2148 pistachio varieties, 1232 of which are Kırmızı and 916 of which are Siirt. In order to classify these images, features of the images were obtained with wavelet image scattering and DarkNet53 convolutional neural network architecture, and then these features were classified with Support Vector Machines (SVM). By using wavelet image scattering and DarkNet53 ESA architecture, 97.98% accuracy was obtained as a result of the classification of the feature set of the images by SVM.

Keywords

• Red variety
• Siirt variety
• Wavelet image scattering
• DarkNet53
• Support vector machines

Kaynakça

1. Acar, I., Eti, S., 2009. Nut Quality of ’Kirmizi’,’Siirt’ and ’Ohadi’ pistachio Cultivars as Affected by Different Pollinators. In V International Symposium on Pistachios and Almonds 912, 81–86. https://doi.org/10.17660/ActaHortic.2011.912.9
2. Akboğa, A., Pakyürek, M., 2020. Farmer Behaviours in Pistachio Growing in Siirt. ISPEC Journal of Agricultural Sciences 4 (2): 36–50.
3. Ataş, M., Doğan,Y., 2015. Classification of Closed and Open Shell Pistachio Nuts by Machine Vision. International Conference on Advanced Technology Sciences, Antalya.
4. Atli, H.S., Arpaci, S., Uyger, N., 2005. Selection of Pistachio Pollinators in Turkey. In IV International Symposium on Pistachios and Almonds 726, 417–22. https://doi.org/10.17660/ActaHortic.2006.726.67.
5. Balta, F., 2002. Phenotypic Differences of Nut and Yield Characteristics in’Siirt Pistachios (Pistacia vera L.) Growth in Siirt Province. Journal Of The American Pomological Society 56 (1): 50.
6. Başaran, E., Cömert, Z., Celik,Y., 2021. Timpanik Membran Görüntü Özellikleri Kullanılarak Sınıflandırılması. Fırat Üniversitesi Mühendislik Bilimleri Dergisi 33 (2): 441–53. https://doi.org/10.35234/fumbd.863118.
7. Basaran, P., Meltem, O., 2009. Occurrence of Aflatoxins in Various Nuts Commercialized in Turkey. Journal of Food Safety 29 (1): 95–105. https://doi.org/10.1111/j.1745-4565.2008.00143.x.
8. Bruna, J., Stéphane, M., 2013. Invariant Scattering Convolution Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 35 (8): 1872–86. https://doi.org/10.1109/TPAMI.2012.230.

9. Çağlar, A., Oktay, T., Hülya, V., Elif, E., 2017. Antepfıstığı (Pistacia vera L.) ve Insan Sağlığı Üzerine Etkileri. Akademik Gıda 15 (4): 436–47. https://doi.org/10.24323/akademik-gida.370408.
10. Cetin, A. E., Pearson, T.C., Tewfik, A.H., 2004. Classification of Closed-and Open-Shell Pistachio Nuts Using Voice-Recognition Technology. Transactions of the ASAE 47 (2): 659.
11. Demir, K., Tümen,V., 2021. Drone-Assisted Automated Plant Diseases Identification Using Spiking Deep Conventional Neural Learning. AI Communications 34: 147–62. https://doi.org/10.3233/AIC-210009.
12. Everest, T., 2021. Suitable Site Selection for Pistachio (Pistacia vera) by Using GIS and Multi-Criteria Decision Analyses (a Case Study in Turkey). Environment, Development and Sustainability 23 (5): 7686–7705. https://doi.org/10.1007/s10668-020-00941-5.
13. Fadaei, H., Suzuki, R., Avtar, R., 2012. Estimation Tree Density as Object-Based in Arid and Semi-Arid Regions Using ALOS. Proceedings of the 4th GEOBIA, 668.
14. Farazi, M., Mohammad, J.A.Z., Hadi, Mi., 2017. A Machine Vision Based Pistachio Sorting Using Transferred Mid-Level Image Representation of Convolutional Neural Network. In 2017 10th Iranian Conference on Machine Vision and Image Processing (MVIP), 145–48. IEEE. https://doi.org/10.1109/IranianMVIP.2017.8342335.
15. Khan, M.A., Alqahtani, A., Khan, A., Shtwai, A., Binbusayyis, A., Ch, A.A., Yong, H., Jaehyuk, C., 2022. Cucumber Leaf Diseases Recognition Using Multi Level Deep Entropy-ELM Feature Selection. Applied Sciences. https://doi.org/10.3390/app12020593.
16. Küden, A.B., Kaska, N., Tanriver, E., Tekin, H., Ak, B.E., 1994. “Determining the Chilling Requirements and Growing Degree Hours of Some Pistachio Nut Cultivars and Regions.” In I International Symposium on Pistachio 419, 85–90. https://doi.org/10.17660/ActaHortic.1995.419.12.
17. Leemans, V., Magein, H., Destain, H., 2002. AE—Automation and Emerging Technologies: On-Line Fruit Grading According to Their External Quality Using Machine Vision. Biosystems Engineering 83 (4): 397–404. https://doi.org/10.1006/bioe.2002.0131.
18. Mart, C., Erkilic, L., Bolu, H., Uygun, N., Altin, M., 1994. Species and Pest Control Methods Used in Pistachio Orchards of Turkey. In I International Symposium on Pistachio 419, 379–86. https://doi.org/10.17660/ActaHortic.1995.419.63.
19. Mei, N., Wang, H., Zhang, Y., Liu, Y., Jiang, X., Wei, S., 2021. Classification of Heart Sounds Based on Quality Assessment and Wavelet Scattering Transform.” Computers in Biology and Medicine 137: 104814. https://doi.org/10.1016/j.compbiomed.2021.104814.
20. Nezhad, R., Ebrahımy, F., 2014. An Intelligent-Based Mechatronics System for Grading the Iranian’s Export Pistachio Nuts into Hulled and Non-Hulled Groups. Indian Journal of ScientificResearch 7 (1): 1063–71.
21. Omid, M., Firouz, M.S., Nouri-Ahmadabadi, H., Mohtasebi, S.S., 2017. Classification of Peeled Pistachio Kernels Using Computer Vision and Color Features. Engineering in Agriculture, Environment and Food 10 (4): 259–65. https://doi.org/10.1016/j.eaef.2017.04.002.
22. Onay, A., Jeffree, C.E., Theobald, C., Yeoman, M.M., 2000. Analysis of the Effects of Maturation Treatments on the Probabilities of Somatic Embryo Germination and Plantlet Regeneration in Pistachio Using a Linear Logistic Method. Plant Cell, Tissue and Organ Culture 60 (2): 121–29. https://doi.org/10.1023/A:1006464505072.
23. Khaleel, R.M., 2019. Feature Detection and Classification of Pistachio by Using Image Processing. (Yükseklisans Tezi) Gaziantep Üniversitesi Fen Bilimleri Enstitüsü Makine Mühendisliği Anabilim Dalı, Gaziantep, 82s.
24. Redmon, J., Farhadi,A., 2018. YOLOv3: An Incremental Improvement. https://doi.org/10.48550/arXiv.1804.02767.
25. Ren, S., Kaiming, H., Girshick, R., Sun, J., 2015. Faster R-Cnn: Towards Real-Time Object Detection with Region Proposal Networks. Advances in Neural Information Processing Systems 28.
26. Shah, F.A., Khan,M.A., Sharif, M., Tariq, U., Khan, A., Kadry, S., Thinnukool, O., 2022. A Cascaded Design of Best Features Selection for Fruit Diseases Recognition. Computers, Materials \& Continua. https://doi.org/10.32604/cmc.2022.019490.
27. Sifre, L., Mallat, S., 2013. Rotation, Scaling and Deformation Invariant Scattering for Texture Discrimination. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1233–40.
28. Simões, A.S., Reali, A.H., Hirakawa, C.A.R., Saraiva, A.M., 2002. Applying Neural Networks to Automated Visual Fruit Sorting. In World Congress of Computers in Agriculture and Natural Resources, Proceedings of the 2002 Conference, 1. American Society of Agricultural and Biological Engineers.
29. Singh, D., Taspinar, Y.S., Kursun, R., Cinar, I., Koklu, M., Ozkan, I.A., Lee, H.N., 2022. Classification and Analysis of Pistachio Species with Pre-Trained Deep Learning Models. Electronics 2022, 11, 981. https://doi.org/10.3390/electronics11070981.
30. Şimşek, M., 2018. “Production Potential and Development Opportunities of Pistachio (Pistacia vera L.) Grown in Southeastern Turkey.” Iğdır Üniversitesi Fen Bilimleri Enstitüsü Dergisi 8 (1): 19–22.
31. Toğaçar, M., Ergen, B., Cömert, Z., 2020. Detection of Lung Cancer on Chest CT Images Using Minimum Redundancy Maximum Relevance Feature Selection Method with Convolutional Neural Networks. Biocybernetics and Biomedical Engineering 40 (1): 23–39. https://doi.org/10.1016/j.bbe.2019.11.004
32. Tsung-Yi, L., Dollár, P., Girshick, R., He, K., Hariharan, B., Belongie, S., 2017. Feature Pyramid Networks for Object Detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2117–25.
33. Unay, D., Gosselin, B., 2005. Artificial Neural Network-Based Segmentation and Apple Grading by Machine Vision. In IEEE International Conference on Image Processing 2005, 2:II--630. IEEE. https://doi.org/10.1109/ICIP.2005.1530134
34. Vapnik, V., 1998. The Support Vector Method of Function Estimation. In Nonlinear Modeling, 55–85. https://doi.org./10.1007/978-1-4615-5703-6_3.
35. Wang, H., Lizhong, D., Hao, Z., Luo, L., Guichao L., Wu, J., Tang, Y., 2021. YOLOv3-Litchi Detection Method of Densely Distributed Litchi in Large Vision Scenes. Edited by Akhil Garg. Mathematical Problems in Engineering 2021: 8883015. https://doi.org/10.1155/2021/8883015.