Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches

Anselme Atchogou; Cengiz Tepe

doi:10.31466/kfbd.1714811

TR EN

Otonom Tarım Aracı İçin Geliştirilmiş Ekin Sırası Tespit Teknikleri: Klasik ve Derin Öğrenme Yaklaşımlarının Karşılaştırmalı Analizi

Abstract

Hassas tarım, küresel nüfus artışı karşısında sürdürülebilir tarım için kritik öneme sahiptir. Bu çalışma, otonom navigasyon için üç ekin sırası tespit yöntemini karşılaştırır: klasik görüntü işleme, CNN tabanlı bitki algılama ve CNN tabanlı ekin sırası segmentasyonu. Yöntemler, doğruluk, hız, ortalama açısal hata ve çevresel uyarlanabilirlik açısından aynı koşullar altında kontrollü bir Gazebo simülasyonunda değerlendirildi. CNN tabanlı ekin sırası segmentasyonu (YOLOv11n-seg), varyasyonlara karşı minimum duyarlılık ile en yüksek doğruluğu (99,5%) elde etti, ancak daha düşük hızlarda çalıştı (ortalama FPS < 5). Klasik görüntü işleme en hızlıydı (ortalama FPS: 95,82), ancak kamera yönelimine ve renk değişikliklerine duyarlılığı nedeniyle daha az güvenilirdi. CNN tabanlı bitki algılama, özellikle YOLOv11n, yüksek doğruluk (98,79%), gerçek zamanlı hız (Jetson Orin Nano'da 32,3 FPS) ve sağlamlık arasında denge kurarak MobileNetV2'yi (94,62%, 21,49 FPS) geride bıraktı. Çalışmanın yeniliği, navigasyon kararlılığını değerlendirmek için ortalama açısal hata kullanılmasıdır. CNN tabanlı yöntemler, özellikle YOLOv11n, klasik yöntemlere (±9,82°) göre daha düşük açısal hatalar (±1,40°) ve daha yüksek kararlılık elde ederek güvenilir simüle edilmiş performansı destekledi. Son zamanlarda yapılan çalışmalar, yöntemlerimizin verimliliğini karşılaştırmalı olarak değerlendirdi. Sonuçlar, gerçek zamanlı kullanım ve doğruluk arasındaki dengeyi vurgulamakta olup, gerçek dünya performansını doğrulamak için saha denemeleri planlanmaktadır.

Keywords

Mahsul sırası algılama, Otonom navigasyon, Derin öğrenme, Ortalama açısal hata

Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches

Abstract

As the global population increases, precision agriculture plays a key role in sustainable farming. This study compares three crop row detection methods to help with autonomous navigation in agriculture: classical image processing, CNN-based plant detection, and CNN-based crop row segmentation. Each method was tested in a controlled Gazebo simulation for accuracy, speed, mean angular error, and adaptability. The CNN-based crop row segmentation method (YOLOv11n-seg) had the highest accuracy (99.5%) and was less affected by environmental changes, but it was slower, with an average speed below 5 FPS. Classical image processing was the fastest (average 95.82 FPS), but it was less reliable because it was sensitive to camera angle and color changes. CNN-based plant detection, especially YOLOv11n, provided a good balance of high accuracy (98.79%), real-time speed (32.3 FPS on Jetson Orin Nano), and robustness, and it performed better than MobileNetV2 (94.62%, 21.49 FPS). The study also used mean angular error to measure navigation stability. CNN-based methods, especially YOLOv11n, had lower angular errors (±1.40°) and were more stable than classical methods (±9.82°), which led to more reliable simulated results. Other recent studies have also compared the efficiency of these methods. The findings highlight a trade-off between real-time performance and accuracy. Field trials are planned to test these results in real-world conditions.

Keywords

Crop row detection, Autonomous navigation, Deep learning, Mean angular error

Supporting Institution

Ondokuz Mayıs University

Project Number

BAP08-2024-5409

Ethical Statement

This study does not involve human or animal subjects and therefore does not require ethical approval.

Thanks

This research was supported by Ondokuz Mayıs University Scientific Research Projects Coordination Unit under the project number BAP08-2024-5409. We thank the university for its financial support.

References

Ahmadi, A., Nardi, L., Chebrolu, N., & Stachniss, C. (2020). Visual servoing-based navigation for monitoring row-crop fields. 2020 IEEE International Conference on Robotics and Automation (ICRA), 4920–4926. https://doi.org/10.1109/ICRA40945.2020.9197114
Alqahtani, D. K., Cheema, M. A., & Toosi, A. N. (2025). Benchmarking deep learning models for object detection on edge computing devices. In W. Gaaloul, Q. Yu, M. Sheng, & S. Yangui (Eds.), Service-Oriented Computing: ICSOC 2024 (Lecture Notes in Computer Science, Vol. 15404, pp. 142–150). Springer. https://doi.org/10.1007/978-981-96-0805-8_11
Bah, M. D., Hafiane, A., & Canals, R. (2020). CRowNet: Deep network for crop row detection in UAV images. IEEE Access, 8, 5189–5200. https://doi.org/10.1109/ACCESS.2019.2960873
Balakirsky, S., & Kootbally, Z. (2012). USARSim/ROS: A combined framework for robotic control and simulation. Proceedings of the ASME/ISCIE 2012 International Symposium on Flexible Automation (ISFA 2012), 101–108. https://doi.org/10.1115/ISFA2012-7179
Bengochea-Guevara, J. M., Conesa-Muñoz, J., Andújar, D., & Ribeiro, A. (2016). Merge fuzzy visual servoing and GPS-based planning to obtain a proper navigation behavior for a small crop-inspection robot. Sensors, 16(3), Article 276. https://doi.org/10.3390/s16030276
Bonadies, S., & Gadsden, S. A. (2019). An overview of autonomous crop row navigation strategies for unmanned ground vehicles. Engineering in Agriculture, Environment and Food, 12(1), 24–31. https://doi.org/10.1016/j.eaef.2018.10.002
Cheppally, R. H., & Sharda, A. (2025). RowDetr: End-to-end crop row detection using polynomials. Smart Agricultural Technology, 7, Article 101494. https://arxiv.org/abs/2412.10525
De Silva, R., Cielniak, G., & Gao, J. (2024). Vision-based crop row navigation under varying field conditions in arable fields. Computers and Electronics in Agriculture, 217, Article 108581. https://doi.org/10.1016/j.compag.2023.108581
De Silva, R., Cielniak, G., Wang, G., & Gao, J. (2023). Deep learning-based crop row detection for infield navigation of agri-robots. Journal of Field Robotics, 40(8), 2299–2321. https://doi.org/10.1002/rob.22238
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009). ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 248–255. https://doi.org/10.1109/CVPR.2009.5206848

Ganapathy, M. R., Periasamy, V., & Ponnusamy, V. (2025). YOLOv11n for precision agriculture: Lightweight and efficient detection of guava defects across diverse conditions. Journal of the Science of Food and Agriculture. https://doi.org/10.1002/jsfa.14331 (In press)
Guo, Z., Quan, L., Sun, D., Lou, Z., Geng, Y., Chen, T., Xue, Y., Hou, P., Wang, C., Wang, J., & He, J. (2024). Efficient crop row detection using transformer-based parameter prediction. Biosystems Engineering, 246, 13–25. https://doi.org/10.1016/j.biosystemseng.2024.07.016
Jegham, N., Koh, S. L., & Alouane, M. H. (2024). YOLO evolution: A comprehensive benchmark and architectural review of YOLOv12, YOLOv11, and their previous versions. arXiv preprint arXiv:2411.00201. https://arxiv.org/abs/2411.00201
Ji, R., & Qi, L. (2011). Crop-row detection algorithm based on Random Hough Transformation. Mathematical and Computer Modelling, 54(3–4), 1016–1020. https://doi.org/10.1016/j.mcm.2010.11.038
Khanam, R., & Hussain, M. (2024). YOLOv11: An overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725. https://arxiv.org/abs/2410.17725
Koenig, N., & Howard, A. (2004). Design and use paradigms for Gazebo, an open-source multi-robot simulator. 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2149–2154. https://doi.org/10.1109/IROS.2004.1389727
Li, W. (2017). Research on the method of generating visual navigation path of kiwi picking robot [Doctoral dissertation, Northwest A&F University]. Northwest A&F University, Yangling, China.
Liao, P.-S., Chen, T.-S., & Chung, P.-C. (2001). A fast algorithm for multilevel thresholding. Journal of Information Science and Engineering, 17(5), 713–727. https://doi.org/10.6688/JISE.2001.17.5.1
Liu, H., Zeng, X., Shen, Y., Xu, J., & Khan, Z. (2025). A single-stage navigation path extraction network for agricultural robots in orchards. Computers and Electronics in Agriculture, 229, Article 109687. https://doi.org/10.1016/j.compag.2024.109687 (In press)
Ma, C., Dong, Z.-Y., Chen, Z.-J., Zhu, Y., & Shi, F. (2021). Research on navigation line generation of kiwi orchard between rows based on root point substitution. Agricultural Research in the Arid Areas, 39(5), 222–230.
Food and Agriculture Organization of the United Nations. (2024). Global report on food crises 2024. FAO. https://doi.org/10.4060/cd3084en (Accessed January 5, 2025, from https://www.fao.org/newsroom/detail/global-report-on-food-crises-is-a-wake-up-call/en)
Oliveira, L. F., Moreira, A. P., & Silva, M. F. (2021). Advances in agriculture robotics: A state-of-the-art review and challenges ahead. Robotics, 10(2), Article 52. https://doi.org/10.3390/robotics10020052
Panda, S. K., Lee, Y., & Jawed, M. K. (2023). Agronav: Autonomous navigation framework for agricultural robots and vehicles using semantic segmentation and semantic line detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 6271–6280. https://doi.org/10.1109/CVPRW59228.2023.00627
Qin, J., Shen, X., Mei, F., & Fang, Z. (2019). An Otsu multi-thresholds segmentation algorithm based on improved ACO. The Journal of Supercomputing, 75(2), 955–967. https://doi.org/10.1007/s11227-018-2622-0
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018). MobileNetV2: Inverted residuals and linear bottlenecks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 4510–4520. https://doi.org/10.1109/CVPR.2018.00474
Shafi, U., Mumtaz, R., García-Nieto, J., Hassan, S. A., Zaidi, S. A. R., & Iqbal, N. (2019). Precision agriculture techniques and practices: From considerations to applications. Sensors, 19(17), Article 3796. https://doi.org/10.3390/s19173796
Shi, J., Bai, Y., Diao, Z., Zhou, J., Yao, X., & Zhang, B. (2023). Row detection-based navigation and guidance for agricultural robots and autonomous vehicles in row-crop fields: Methods and applications. Agronomy, 13(7), Article 1780. https://doi.org/10.3390/agronomy13071780
Tan, Y., Su, W., Zhao, L., Lai, Q., Wang, C., Jiang, J., Wang, Y., & Li, P. (2023). Navigation path extraction for inter-row robots in Panax notoginseng shade house based on Im-YOLOv5s. Frontiers in Plant Science, 14, Article 1246717. https://doi.org/10.3389/fpls.2023.1246717
Vidović, I., Cupec, R., & Hocenski, Ž. (2016). Crop row detection by global energy minimization. Pattern Recognition, 55, 68–86. https://doi.org/10.1016/j.patcog.2016.01.013
Yang, R., Zhai, Y., Zhang, J., Zhang, H., Tian, G., Zhang, J., Huang, P., & Li, L. (2022a). Potato visual navigation line detection based on deep learning and feature midpoint adaptation. Agriculture, 12(9), Article 1363. https://doi.org/10.3390/agriculture12091363
Yang, Z., Ouyang, L., Zhang, Z., Duan, J., Yu, J., & Wang, H. (2022b). Visual navigation path extraction of orchard hard pavement based on scanning method and neural network. Computers and Electronics in Agriculture, 197, Article 106964. https://doi.org/10.1016/j.compag.2022.106964
Zhang, S., Liu, Y., Xiong, K., Tian, Y., Du, Y., Zhu, Z., & Zhai, Z. (2024). A review of vision-based crop row detection method: Focusing on field ground autonomous navigation operations. Computers and Electronics in Agriculture, 222, Article 109086. https://doi.org/10.1016/j.compag.2024.109086
Zheng, Z., Hu, Y., Li, X., & Huang, Y. (2023). Autonomous navigation method of jujube catch-and-shake harvesting robot based on convolutional neural networks. Computers and Electronics in Agriculture, 215, Article 108469. https://doi.org/10.1016/j.compag.2023.108469
Zhou, J., Geng, S., Qiu, Q., Shao, Y., & Zhang, M. (2022). A deep-learning extraction method for orchard visual navigation lines. Agriculture, 12(10), Article 1650. https://doi.org/10.3390/agriculture12101650

Details

Primary Language

English

Subjects

Precision Agriculture Technologies

Journal Section

Research Article

Authors

Anselme Atchogou ^*
0009-0002-1593-516X
Benin

Cengiz Tepe
0000-0003-4065-5207
Türkiye

Publication Date

May 11, 2026

Submission Date

June 5, 2025

Acceptance Date

December 12, 2025

Published in Issue

Year 2026 Volume: 16 Number: 2

DOI

https://doi.org/10.31466/kfbd.1714811

IZ

https://izlik.org/JA38WB74EF

APA

Atchogou, A., & Tepe, C. (2026). Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches. Karadeniz Fen Bilimleri Dergisi, 16(2), 605-625. https://doi.org/10.31466/kfbd.1714811

AMA

1.Atchogou A, Tepe C. Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches. KFBD. 2026;16(2):605-625. doi:10.31466/kfbd.1714811

Chicago

Atchogou, Anselme, and Cengiz Tepe. 2026. “Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches”. Karadeniz Fen Bilimleri Dergisi 16 (2): 605-25. https://doi.org/10.31466/kfbd.1714811.

EndNote

Atchogou A, Tepe C (May 1, 2026) Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches. Karadeniz Fen Bilimleri Dergisi 16 2 605–625.

IEEE

[1]A. Atchogou and C. Tepe, “Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches”, KFBD, vol. 16, no. 2, pp. 605–625, May 2026, doi: 10.31466/kfbd.1714811.

ISNAD

Atchogou, Anselme - Tepe, Cengiz. “Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches”. Karadeniz Fen Bilimleri Dergisi 16/2 (May 1, 2026): 605-625. https://doi.org/10.31466/kfbd.1714811.

JAMA

1.Atchogou A, Tepe C. Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches. KFBD. 2026;16:605–625.

MLA

Atchogou, Anselme, and Cengiz Tepe. “Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches”. Karadeniz Fen Bilimleri Dergisi, vol. 16, no. 2, May 2026, pp. 605-2, doi:10.31466/kfbd.1714811.

Vancouver

1.Anselme Atchogou, Cengiz Tepe. Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches. KFBD. 2026 May 1;16(2):605-2. doi:10.31466/kfbd.1714811

The published articles in The Black Sea Journal of Sciences are licensed under Creative Commons Attribution-NonCommercial 4.0 International

Otonom Tarım Aracı İçin Geliştirilmiş Ekin Sırası Tespit Teknikleri: Klasik ve Derin Öğrenme Yaklaşımlarının Karşılaştırmalı Analizi

Abstract

Keywords

Enhanced Crop Row Detection Techniques for Autonomous Agricultural Vehicle: A Comparative Analysis of Classical and Deep Learning Approaches

Abstract

Keywords

Supporting Institution

Project Number

Ethical Statement

Thanks

References

Details

Primary Language

Subjects

Journal Section

Authors

Publication Date

Submission Date

Acceptance Date

Published in Issue

DOI

IZ

Cite