YOLOv8 ve RF-DETR Kullanılarak Sarı Yapışkan Tuzaklardaki Böceklerin Yapay Zeka Destekli Tespiti ve Sınıflandırılması: Hassas Tarım için Derin Öğrenme Yaklaşımı

Yıl 2025, Cilt: 41 Sayı: 3, 812 - 823, 31.12.2025

Fatma Öncü , Fehim Köylü

https://doi.org/10.65520/erciyesfen.1748404

https://izlik.org/JA32BT66XM

Öz

Tarımda zararlı böceklerin erken ve doğru tespiti, ürün verimliliğini artırmak ve pestisit kullanımını azaltmak açısından önemlidir. Geleneksel sarı yapışkan ve feromon tuzaklarında yapılan manuel sayımlar iş gücü ve zaman açısından maliyetli olup, hata riski taşır. Özellikle sera ortamında değişken zararlı yoğunluğu nedeniyle bu yöntemler yetersiz kalmaktadır.
Bu çalışmada, dijital feromon tuzaklarından elde edilen yüksek çözünürlüklü görüntüler kullanılarak üç farklı böcek türünün otomatik tespiti ve sınıflandırılması amaçlanmıştır: Trialeurodes vaporariorum (beyaz sinek), Macrolophus pygmaeus ve Nesidiocoris tenuis. Görüntüler sera ortamında toplanmış, uzmanlarca etiketlenmiştir. YOLOv8 ve RF-DETR modelleri ile nesne tespiti yapılmış: tespit edilen böcekler VGG19, ResNet50, NASNet Mobile ve FGCN gibi modellerle sınıflandırılmıştır.
Sonuçlara göre, YOLOv8 küçük nesne tespitinde %81,6 mAP50 başarısı göstermiştir. RF-DETR, büyük nesnelerde %77,1 mAP50 değeriyle daha başarılı olmuştur. Sınıflandırmada ise VGG19 modeli %97.05 doğruluk ile en yüksek performansı sağlamıştır.
Bu çalışma, düşük maliyetli ve yüksek doğruluklu dijital feromon tuzaklarına entegre edilebilecek bir sistem önererek, akıllı tarım uygulamaları için katkı sağlamaktadır.

Anahtar Kelimeler

Derin Öğrenme , Nesne Tespiti , Dijital Feromon Tuzağı , Akıllı Tarım , Sarı Yapışkan Tuzak , Otomatik Böcek Tanıma

Etik Beyan

Bu çalışmadaki tüm bilgilerin, akademik ve etik kurallara uygun bir şekilde elde edildiğini beyan ederim. Aynı zamanda bu kural ve davranışların gerektirdiği gibi, bu çalışmanın özünde olmayan tüm materyal ve sonuçları tam olarak aktardığımı ve referans gösterdiğimi belirtirim.

Destekleyen Kurum

Bu çalışma herhangi bir kurumsal maddi destek alınmaksızın yürütülmüştür.

Teşekkür

Yüksek lisans tez çalışmam süresince bana her daim rehberlik eden, akademik gelişimime büyük katkı sağlayan, bilgi ve tecrübelerini sabırla paylaşan değerli danışmanım Dr. Öğr. Üyesi Fehim KÖYLÜ ’ye en içten teşekkürlerimi sunarım

AI-Powered Detection and Classification of Insects on Yellow Sticky Traps Using YOLOv8 and RF-DETR: Deep Learning for Precision Agriculture

https://izlik.org/JA32BT66XM

Öz

Early and accurate detection of pests in agriculture is crucial for increasing crop yields and reducing pesticide use. Manual counts using traditional yellow sticky traps and pheromone traps are labor-intensive, time-consuming, and prone to error. These methods are particularly inadequate in greenhouse environments due to variable pest density.
This study aimed to automatically detect and classify three different insect species using high-resolution images obtained from digital pheromone traps: Trialeurodes vaporariorum(whitefly), Macrolophus pygmaeus, and Nesidiocoris tenuis. The images were collected in a greenhouse and labeled by experts. Object detection was performed using YOLOv8 and RF-DETR models: the detected insects were classified using models such as VGG19, ResNet50, NasNet Mobile, and FGCN. According to the results, YOLOv8 achieved an 81.6% mAP50 accuracy in small object detection. RF- DETR performed better with a 77.1% mAP50 for larger objects. The VGG19 model achieved the highest classification performance with 97.05% accuracy. This study contributes to smart agriculture applications by proposing a system that can be integrated into low-cost, high accuracy digital pheromone traps.

Anahtar Kelimeler

Deep Learning , Object Detection , Digital Pheromone Trap , Smart Agriculture , Yellow Sticky Trap , Automated Insect Recognition

Kaynakça

• Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Houlsby, N., Gelly, S., Uszkoreit, J., Heigold, G. 2021. An image is worth 16×16 words: Transformers for image recognition at scale. International Conference on Learning Representations (ICLR), https://arxiv.org/abs/2010.11929
• Potamitis, I., Rigakis, I., Fysarakis, K. 2020. Insect detection from acoustic signals using low-cost, low-power devices. Computers and Electronics in Agriculture, 168, 105121, https://doi.org/10.1016/j.compag.2019.105121
• Andreu-Vilar, M. V., Garcia, L., Garcia-Sanchez, A., Asorey-Cacheda, R., Garcia-Haro, J. 2024. Enhancing precision agriculture pest control: A generalized deep learning approach with YOLOv8-based insect detection. IEEE Access, 12: 14567–14575, https://doi.org/10.1109/access.2024.3413979
• Gao, Y., Yin, F., Hong, C., Chen, X., Deng, H., Liu, Y., Li, Z., Yu, Q. 2024. Intelligent field monitoring system for cruciferous vegetable pests using yellow sticky trap images and an improved cascade R-CNN. Journal of Integrative Agriculture, 23(4): 1023–1035, https://doi.org/10.1016/j.jia.2024.06.017
• Zhang, X., Li, Z., Ren, L., Liu, X., Zeng, T., Tao, J. 2024. Detection and recognition of the invasive species, Hylurgus ligniperda, in traps, based on a cascaded convolutional neural network. (Web sayfası: https://onlinelibrary.wiley.com), (Erişim tarihi: 7 Nisan 2024), https://doi.org/10.1002/ps.8126
• Zhu, X., Su, W., Lu, L., Li, B., Wang, X., Dai, J. 2020. Deformable DETR: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159, https://arxiv.org/abs/2010.04159.
• Zhao, X., Shen, L., Yu, Z., Xu, H. 2021. Understanding the limitations of DETR on small object detection. arXiv, https://arxiv.org/abs/2107.04528
• Gao, P., Zheng, M., Wang, X., Dai, J., Li, H. 2021. Fast convergence of DETR with spatially modulated co-attention. arXiv. https://doi.org/10.48550/arXiv.2108.02404
• Rahman, C. R., Arko, P. S., Ali, M. E., Khan, M. A. I., Apon, S. H., Nowrin, F., Wasif, A. 2020. Identification and recognition of rice diseases and pests using convolutional neural networks. Biosystems Engineering, 194, 112–120, https://doi.org/10.1016/j.biosystemseng.2020.03.020
• Liu, H., Jiang, H., Cheng, C., Liu, S. 2023. Efficient pest detection in complex backgrounds using an improved DEYOLO-pest model. Computers and Electronics in Agriculture, 205: 107622.
• Nieuwenhuizen, A. T., Hemming, J., Janssen, D., Suh, H. K., Bosmans, L., Sluydts, V., Brenard, N., Rodríguez, E., Tellez, M. D. M. 2019. Raw data from yellow sticky traps with insects for training of deep learning convolutional neural network for object detection [Dataset]. Wageningen University & Research https://doi.org/10.4121/uuid:8b8ba63a-1010-4de7-a7fb-6f9e3baf128e
• Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S. 2020. End-to-end object detection with transformers. Computer Vision – ECCV 2020: 16th European Conference, Proceedings 213–229, Springer. https://doi.org/10.1007/978-3-030-58452-8_13
• Malathi, V., Gopinath, M. P. 2021. Classification of pest detection in paddy crop based on transfer learning approach. Acta Agriculturae Scandinavica, Section B – Soil, Plant Science, 71(7), 552–559. https://doi.org/10.1080/09064710.2021.1874045
• Deng, Y., Chen, Z., Zhang, X., Fan, X., Shi, Z. 2023. DETRs meet small objects: A survey (arXiv:2306.04643). https://doi.org/10.48550/arXiv.2306.04643
• Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Houlsby, N., Gelly, S., Uszkoreit, J., & Heigold, G. 2021. An image is worth 16×16 words: Transformers for image recognition at scale. International Conference on Learning Representations (ICLR). https://arxiv.org/abs/2010.11929
• Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z. 2016. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2818–2826. https://doi.org/10.1109/CVPR.2016.308
• Zoph, B., Vasudevan, V., Shlens, J., & Le, Q. V. 2018. Learning transferable architectures for scalable image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 8697–8710, IEEE. https://doi.org/10.1109/CVPR.2018.00907
• Liu, Z., Mao, H., Wu, C. Y., Feichtenhofer, C., Darrell, T., Xie, S. 2022. A ConvNet for the 2020s. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 11976–11986. https://doi.org/10.1109/CVPR52688.2022.01170
• Roboflow. 2024. Yellow Sticky Trap Insect Dataset. Roboflow Universe. Retrieved April 2024, from https://universe.roboflow.com
• 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 (CVPR) 4700–4708. https://doi.org/10.1109/CVPR.2017.243
• Andreu-Vilar, M. V., Garcia, L., Garcia-Sanchez, A., Asorey-Cacheda, R., Garcia-Haro, J. 2024. Enhancing precision agriculture pest control: A generalized deep learning approach with YOLOv8-based insect detection. IEEE Access, 12: 14567–14575. https://doi.org/10.1109/access.2024.3413979
• Liu, S., Qi, L., Qin, H., Shi, J., & Jia, J. 2018. Path Aggregation Network for Instance Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
• Tan, M., & Le, Q. V. 2019. EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML) 6105–6114. PMLR. https://arxiv.org/abs/1905.11946
• Jocher, G., Chaurasia, A., Qiu, J., & Stoken, A. 2023. YOLO by Ultralytics [Computer software]. GitHub. https://github.com/ultralytics/ultralytics
• Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020). End-to-end object detection with transformers. In Computer Vision – ECCV 2020: European Conference on Computer Vision 213–229. Springer International Publishing. https://doi.org/10.1007/978-3-030-58452-8_13 26. Tao, Y., Pan, Y., Wu, Y., Chen, Z., Zhou, M. 2020. Real-time monitoring system for greenhouse insects based on embedded system and machine vision. Computers and Electronics in Agriculture, 170, 105247. https://doi.org/10.1016/j.compag.2020.105247
• Liu, S., Qi, L., Qin, H., Shi, J., Jia, J. 2018. Path Aggregation Network for Instance Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
• Zhang, C., Li, H., Liu, J., Wang, Z. 2020. An automatic counting system for aphids on leaves using deep learning. Biosystems Engineering, 189: 168–180.
• Liu, S., Qi, L., Qin, H., Shi, J., Jia, J. 2018. Path Aggregation Network for Instance Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
• Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S. 2020. End-to-end object detection with transformers. In Computer Vision – ECCV 2020: European Conference on Computer Vision 213–229. Springer International Publishing. https://doi.org/10.1007/978-3-030-58452-8_13
• Li, W., Zhao, Y., Liu, Z., Zhang, J., Xu, L. 2019. Automatic detection and classification of pest insects in field using multi-task learning. Computers and Electronics in Agriculture, 167, 105062. https://doi.org/10.1016/j.compag.2019.105062
• Suto, J. 2024. Improving the generalization capability of YOLOv5 on remote sensed insect trap images with data augmentation. Multimedia Tools and Applications, 83(6): 3478–3496. https://doi.org/10.1007/s11042-023-16578-1
• Hacinas, E. A., Querol, L. S., Santos, K. L. T., Matira, E. B., Castillo, R. C., Arcelo, M., Amalin, D., Rustia, D. J. A. 2024. Rapid automatic cacao pod borer detection using edge computing on low-end mobile devices. Agronomy, 14(502). (Web sayfası: https://www.mdpi.com/journal/agronomy), (Erişim tarihi: Nisan 2024). https://doi.org/10.3390/agronomy14030502
• Li, H., Yuan, W., Xia, Y., Wang, Z., He, J., Wang, Q., Zhang, S., Li, L., Yang, F., Wang, B. 2024. YOLOv8n-WSE-Pest: A lightweight deep learning model based on YOLOv8n for pest identification in tea gardens. Applied Sciences, 14(1). (Web sayfası: https://www.mdpi.com/journal/applsci), (Erişim tarihi: Nisan 2024). https://doi.org/10.3390/app14198748
• Cheng, X., Zhang, Y., Chen, Y., Wu, Y., Yue, Y. 2017. Pest identification via deep residual learning in complex background. Computers and Electronics in Agriculture, 141, 351–356. https://doi.org/10.1016/j.compag.2017.08.005
• Li, Y., Wang, H., Dang, L. M., Sadeghi Niaraki, A., Moon, H. 2020. Crop pest recognition in natural scenes using convolutional neural networks. Computers and Electronics in Agriculture, 169, 105174. https://doi.org/10.1016/j.compag.2019.105174
• Santhosh, B. J., Adithya, H. R., Akash Kumar, G. S., Hemanth Gowda, R. M., Jaswanth, H. V. 2024. Utilizing deep learning and image processing for detection of pests. 2024 Second International Conference on Advances in Information Technology (ICAIT), 1–8. IEEE. https://doi.org/10.1109/ICAIT61638.2024.10690519
• Tetila, E. C., Machado, B. B., Astolfi, G., de Souza Belete, N. A., Amorim, W. P., Roel, A. R., Pistori, H. 2020. Detection and classification of soybean pests using deep learning with UAV images. Computers and Electronics in Agriculture, 179, 105836. https://doi.org/10.1016/j.compag.2020.105836
• Pattnaik, G., Shrivastava, V. K., Parvathi, K. 2020. Transfer learning-based framework for classification of pest in tomato plants. Applied Artificial Intelligence, 34(13), 981–993. https://doi.org/10.1080/08839514.2020.1792034