Research Article
BibTex RIS Cite

Fındık Meyvesinin Sınıflandırılmasında Shufflenet ve Densenet Derin Öğrenme Algoritmalarinin Performansının Karşılaştırılması

Year 2025, Volume: 2 Issue: 2, 96 - 102, 30.10.2025
https://doi.org/10.5281/zenodo.17474580

Abstract

Yapay zekâ kullanımıyla fındık meyvesinin sınıflandırılması, verimliliğinin artırılması mümkün olabilir. Bu çalışmadaki amacımız fındık meyvesinin sınıflandırılmasında kullanılabilecek bir otomatik sınıflandırma sistemi için bir yöntem önermektir. Fındık meyvesinin sınıflandırılmasında derin öğrenme algoritmalarının kullanılması, tarım ve gıda endüstrisinde kalite kontrol, otomatik ayıklama ve verimlilik artırma gibi alanlarda önemli bir rol oynayabilir. Yapay zekânın fındık görüntülerini analiz etmesinde kullanılabilecek derin öğrenme algoritmaları incelenmiştir. Shufflenet ve DenseNet derin öğrenme algoritmalarının fındık meyvesine ait veri tabanında performansları karşılaştırılmıştır. Analizler sonucunda ShuffleNet eğitimde % 99.79, test aşamasında % 99.94 doğruluğa ulaşmıştır. DenseNet ise eğitimde % 99.98, test aşamasında % 99.95 doğruluğa ulaşmıştır. Fındık meyvesinin sınıflandırılmasında derin öğrenmenin, insan gözünden daha hızlı ve tutarlı sonuçlar verebiliceği sonucuna ulaşabiliriz.

References

  • [1] I. Khosa and E. Pasero, “Feature extraction in X-ray images for hazelnuts classification,” in Proc. 2014 International Joint Conference on Neural Networks (IJCNN), Beijing, China, 2014, pp. 2354–2360, doi: 10.1109/IJCNN.2014.6889661.
  • [2] S. Bayrakdar, B. Çomak, D. Başol, and İ. Yücedağ, “Determination of type and quality of hazelnut using image processing techniques,” in Proc. 2015 23rd Signal Processing and Communications Applications Conference (SIU), Malatya, Turkey, 2015, pp. 616–619, doi: 10.1109/SIU.2015.7129899.
  • [3] C. Koç, D. Gerdan, M. B. Eminoğlu, U. Yegül, B. Koç, and M. Vatandaş, “Classification of hazelnut cultivars: comparison of DL4J and ensemble learning algorithms,” Notulae Botanicae Horti Agrobotanici Cluj-Napoca, vol. 48, no. 4, pp. 2316–2327, Dec. 2020, doi: 10.15835/nbha48412041.
  • [4] S. Solak and U. Altınışık, “Detection and classification of hazelnut fruit by using image processing techniques and clustering methods,” Sakarya University Journal of Science, vol. 22, no. 1, pp. 56–65, Feb. 2018, doi: 10.16984/saufenbilder.303850.
  • [5] A. Kamilaris and F. X. Prenafeta-Boldú, “Deep learning in agriculture: A survey,” Computers and Electronics in Agriculture, vol. 147, pp. 70–90, Apr. 2018, doi: 10.1016/j.compag.2018.02.016.
  • [6] K. P. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Computers and Electronics in Agriculture, vol. 145, pp. 311–318, Feb. 2018, doi: 10.1016/j.compag.2018.01.009.
  • [7] O. Keles and A. Taner, “Classification of hazelnut varieties by using artificial neural network and discriminant analysis,” Spanish Journal of Agricultural Research, vol. 19, no. 4, e0211, 2021, doi: 10.5424/sjar/2021194-18056.
  • [8] X. Zhang, X. Zhou, M. Lin, and J. Sun, “ShuffleNet: An extremely efficient convolutional neural network for mobile devices,” arXiv preprint, arXiv:1707.01083, Dec. 2017, doi: 10.48550/arXiv.1707.01083.
  • [9] G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” arXiv preprint, arXiv:1608.06993, Jan. 2018, doi: 10.48550/arXiv.1608.06993.
  • [10] E. Güneş, Hazelnuts Dataset (Version 2), Marmara University, Jun. 2022. [Online]. Available: https://doi.org/10.17632/dvvx6kst3f.2.
  • [11] Ö. Tomak, “Elektrokardiyografi sinyallerinde deneysel mod ayrıştırma ve geliştirilmiş karar ağaçları kullanarak aritmi tespiti,” Karadeniz Fen Bilimleri Dergisi, vol. 9, no. 1, pp. 103–110, 2019.

Comparison of The Performance of Shufflenet and Densenet Deep Learning Algorithms in The Classification of Hazelnut Fruit

Year 2025, Volume: 2 Issue: 2, 96 - 102, 30.10.2025
https://doi.org/10.5281/zenodo.17474580

Abstract

It may be possible to increase the efficiency of hazelnut fruit classification by using artificial intelligence. Our aim in this study is to propose a method for an automatic classification system that can be used in the classification of hazelnut fruit. The use of deep learning algorithms in hazelnut fruit classification can play an important role in areas such as quality control, automatic sorting, and efficiency enhancement in the agricultural and food industry. Deep learning algorithms that can be used in the analysis of hazelnut images by artificial intelligence were examined. The performances of Shufflenet and DenseNet deep learning algorithms on the hazelnut fruit database were compared. As a result of the analysis, ShuffleNet achieved 99.79% accuracy in training and 99.94% in the test phase. DenseNet achieved 99.98% accuracy in training and 99.95% in the test phase. We can conclude that deep learning can provide faster and more consistent results than the human eye in the classification of hazelnut fruit.

References

  • [1] I. Khosa and E. Pasero, “Feature extraction in X-ray images for hazelnuts classification,” in Proc. 2014 International Joint Conference on Neural Networks (IJCNN), Beijing, China, 2014, pp. 2354–2360, doi: 10.1109/IJCNN.2014.6889661.
  • [2] S. Bayrakdar, B. Çomak, D. Başol, and İ. Yücedağ, “Determination of type and quality of hazelnut using image processing techniques,” in Proc. 2015 23rd Signal Processing and Communications Applications Conference (SIU), Malatya, Turkey, 2015, pp. 616–619, doi: 10.1109/SIU.2015.7129899.
  • [3] C. Koç, D. Gerdan, M. B. Eminoğlu, U. Yegül, B. Koç, and M. Vatandaş, “Classification of hazelnut cultivars: comparison of DL4J and ensemble learning algorithms,” Notulae Botanicae Horti Agrobotanici Cluj-Napoca, vol. 48, no. 4, pp. 2316–2327, Dec. 2020, doi: 10.15835/nbha48412041.
  • [4] S. Solak and U. Altınışık, “Detection and classification of hazelnut fruit by using image processing techniques and clustering methods,” Sakarya University Journal of Science, vol. 22, no. 1, pp. 56–65, Feb. 2018, doi: 10.16984/saufenbilder.303850.
  • [5] A. Kamilaris and F. X. Prenafeta-Boldú, “Deep learning in agriculture: A survey,” Computers and Electronics in Agriculture, vol. 147, pp. 70–90, Apr. 2018, doi: 10.1016/j.compag.2018.02.016.
  • [6] K. P. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Computers and Electronics in Agriculture, vol. 145, pp. 311–318, Feb. 2018, doi: 10.1016/j.compag.2018.01.009.
  • [7] O. Keles and A. Taner, “Classification of hazelnut varieties by using artificial neural network and discriminant analysis,” Spanish Journal of Agricultural Research, vol. 19, no. 4, e0211, 2021, doi: 10.5424/sjar/2021194-18056.
  • [8] X. Zhang, X. Zhou, M. Lin, and J. Sun, “ShuffleNet: An extremely efficient convolutional neural network for mobile devices,” arXiv preprint, arXiv:1707.01083, Dec. 2017, doi: 10.48550/arXiv.1707.01083.
  • [9] G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” arXiv preprint, arXiv:1608.06993, Jan. 2018, doi: 10.48550/arXiv.1608.06993.
  • [10] E. Güneş, Hazelnuts Dataset (Version 2), Marmara University, Jun. 2022. [Online]. Available: https://doi.org/10.17632/dvvx6kst3f.2.
  • [11] Ö. Tomak, “Elektrokardiyografi sinyallerinde deneysel mod ayrıştırma ve geliştirilmiş karar ağaçları kullanarak aritmi tespiti,” Karadeniz Fen Bilimleri Dergisi, vol. 9, no. 1, pp. 103–110, 2019.
There are 11 citations in total.

Details

Primary Language Turkish
Subjects Electrical Engineering (Other)
Journal Section Research Article
Authors

Özgür Tomak 0000-0003-2993-6913

Publication Date October 30, 2025
Submission Date June 11, 2025
Acceptance Date October 1, 2025
Published in Issue Year 2025 Volume: 2 Issue: 2

Cite

IEEE Ö. Tomak, “Fındık Meyvesinin Sınıflandırılmasında Shufflenet ve Densenet Derin Öğrenme Algoritmalarinin Performansının Karşılaştırılması”, HENDESE, vol. 2, no. 2, pp. 96–102, 2025, doi: 10.5281/zenodo.17474580.