Improvement and Implementation of a Potato Recognition Algorithm Based on YOLOv8

Dongxuan Huang; Mingge Sun; Sen Ye; Jiaxuan Chai

doi:10.54097/8jm78164

Authors

Dongxuan Huang
Mingge Sun
Sen Ye
Jiaxuan Chai

DOI:

https://doi.org/10.54097/8jm78164

Keywords:

Deep Learning, YOLOv8, Attention Mechanism, Potato Recognition

Abstract

To address issues such as poor recognition accuracy and low efficiency caused by the large number of potato targets that need to be processed at once during potato identification, this paper proposes an improved YOLOv8-based potato recognition algorithm. The improvements include: introducing DynamicConv in the backbone network to enhance model representation capability; replacing the C2f module in the backbone network with C2f_SAConv to strengthen feature extraction in object detection and segmentation tasks; incorporating Slim-neck in the Neck layer to reduce model parameters and floating-point operations; and adding an EMA attention mechanism to improve the model's focus on different potatoes. The improved YOLOv8 algorithm achieves an average recognition accuracy of 89.7%, reduces computational load by 27.2%, and facilitates deployment on resource-constrained embedded devices.

Downloads

Download data is not yet available.

References

[1] Tang Jianzhao, Wang Jing, Xiao Dengpan, et al. Research Progress and Development Prospects of Potato Growth Models [J]. Scientia Agricultura Sinica,2021,54(05):921-932.

[2] Hao Ruoshi, Lyu Jianfei, Wang Aoxue, et al. New Characteristics and Development Trends of China's Potato Industry Under the Background of High-Quality Development [J]. Agricultural Outlook,2024,20(01):7-12.

[3] Li Zhonghui, Zhang Haocheng, Wang Xiuli. Current Status and Development Prospects of Potato Supply and Demand in China [J]. Agricultural Outlook,2022, 18(06): 79-85.

[4] Liang Jurong, Huang Biao, Tian Haijiang, et al. Research Status of Automatic Potato Grading Machines [J]. Hans Journal of Agricultural Sciences, 2021, 11: 138.

[5] Li Qi, Wang Jun. Potato Grading Method Based on Fusion of Color and Texture Features[J]. Science Technology and Engineering, 2019, 19(25): 273-279.

[6] Al-Kateb G, Mijwil M M, Aljanabi M, et al. AI-PotatoGuard: Leveraging Generative Models for Early Detection of Potato Diseases [J]. Potato Research, 2024: 1-15.

[7] Liu Yijun. Research on Potato External Quality Grading Method Based on Machine Vision [D]. Chinese Academy of Agricultural Mechanization Sciences, DOI: 10.27629/d. cnki. gznjk. 2023.000009.

[8] Akther J, Nayan A A, Harun-Or-Roshid M. Potato leaves blight disease recognition and categorization using deep learning[J]. Engineering Journal, 2023, 27(9): 27-38.

[9] ZHANG W, Yuelin H A N, Huang C, et al. Recognition method for seed potato buds based on improved YOLOv3-tiny[J]. INMATEH-Agricultural Engineering, 2022, 67(2).

[10] Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023: 7464-7475.

[11] C G S, Upadhyay A, Zhang Y, et al.Field-based multispecies weed and crop detection using ground robots and advanced YOLO models: A data and model-centric approach[J].Smart Agricultural Technology, 2024, 9100538-100538.

[12] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. Springer International Publishing, 2016: 21-37.

[13] Shih K H, Chiu C T, Lin J A, et al. Real-time object detection with reduced region proposal network via multi-feature concatenation[J]. IEEE transactions on neural networks and learning systems, 2019, 31(6): 2164-2173.

[14] Zhu X, Lyu S, Wang X, et al. TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]// Proceedings of the IEEE/CVF international conference on computer vision. 2021: 2778-2788.

[15] Wang A, Chen H, Liu L, et al. Yolov10: Real-time end-to-end object detection[J]. arXiv preprint arXiv:2405. 14458, 2024.