Е-Публикации
Технически университет - София

Abstract: In the resent years machine learning has being applied to various arias of agriculture. Machine learning (ML) plays a crucial role in addressing weed-related issues in agriculture. ML algorithms, particularly those based on deep learning, can learn intricate patterns and features, making them well-suited for the nuanced task of weed detection. In the present research deep learning algorithms such as YOLOv5 and SORT are adapted and applied to detect and to track vegetables and weeds based on computer vision. First, datasets of annotated full color images of cabbage and weed are created. Then, the ML algorithm YOLOv5 for object detection is selected. It detects and localize objects in a frame in real time from video. Next, YOLOv5 is trained to detect cabbage and weed form our annotated datasets. The accuracy of the model performance is studied. The real-time tracking algorithm SORT is selected and adjusted to track cabbages and weeds from videos. SORT is a high – speed, accurate and robust tracking algorithm. Moreover, the object tracker SORT is adjusted to assign to each vegetable and weed unique number. Finally, the modified object detector and tracker provide significant information about the distribution, sizes and numbers of weeds and vegetables in the given areas. This lead to efficient weed treatment procedure.

References

1. Kubiak, A., Wolna-Maruwka, A., Niewiadomska, A., Pilarska, A.A.: The problem of weed infestation of agricultural plantations vs. the assumptions of the european biodiversity strategy. Agronomy 12, 1808 (2022). https://doi.org/10.3390/agronomy12081808
2. Gonzalez-Andujar, J.L.: Integrated weed management: a shift towards more sustainable and holistic practices. Agronomy 13, 2646 (2023). https://doi.org/10.3390/agronomy13102646
3. Brand, J., Yaduraju, N.T., Shivakumar, B.G., Murray, L.: Weed management. In: Yadav, S.S., McNeil, D.L., Stevenson, P.C. (eds.) Lentil. Springer, Dordrecht (2007). https://doi.org/10.1007/978-1-4020-6313-8_10
4. Fetahaj, R., Kabashi, B., Mehmeti, A.: Parasitic plants in forage legumes – Medicago sativa L., parasitic plants. IntechOpen (2022). https://doi.org/10.5772/intechopen.100539
5. Khurana, G., Bawa, N.K.: Weed detection approach using feature extraction and KNN classification. In: Pandey, V.C., Pandey, P.M., Garg, S.K. (eds.) Advances in Electromechanical Technologies. Lecture Notes in Mechanical Engineering. Springer, Singapore (2021). https://doi.org/10.1007/978-981-15-5463-6_93
6. Uddin, M., Bansal, J.: Computer vision and machine learning in agriculture. Springer Singapore https://doi.org/10.1007/978-981-33-6424-0. eBook ISBN 978-981-33-6424-0
7. Benos, L., Tagarakis, A.C., Dolias, G., Berruto, R., Kateris, D., Bochtis, D.: Machine learning in agriculture: a comprehensive updated review. Sensors 21(11), 3758 (2021). https://doi.org/10.3390/s21113758
8. Meshram, V., Patil, K., Meshram, V., Hanchate, D., Ramkteke, S.D.: Machine learning in agriculture domain: a state-of-art survey. Artif. Intell. Life Sci. 1 (2021). https://doi.org/10.1016/j.ailsci.2021.100010
9. Habib, M.T., Raza, D.M., Islam, M.M., Victor, D.B., Arif, M.A.I.: Applications of computer vision and machine learning in agriculture: a state-of-the-art glimpse. In: International Conference on Innovative Trends in Information Technology (ICITIIT), Kottayam, India, 2022, pp. 1-5 (2022). https://doi.org/10.1109/ICITIIT54346.2022.9744150
10. Magalhães, S.A., et al.: Evaluating the single-shot MultiBox detector and YOLO deep learning models for the detection of tomatoes in a greenhouse. Sensors 21(10), 3569 (2021). https://doi.org/10.3390/s21103569
11. Umamaheswari, S., Arjun, R., Meganathan, D.: Weed detection in farm crops using parallel image processing. In: Conference on Information and Communication Technology (CICT), Jabalpur, India (2018)
12. Toskova, A., Toskov, B.: A deep learning wheat-weed dataset. In: International Conference Automatics and Informatics (ICAI), Varna, Bulgaria 2023, pp. 502– 507 (2023). https://doi.org/10.1109/ICAI58806.2023.10339055
13. Louridas, P., Ebert, C.: Machine learning. IEEE Softw. 33(5), 110–115 (2016). https://doi.org/10.1109/MS.2016.114
14. Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, pp. 779–788 (2016)
15. Liu, W., et al.: SSD: single shot MultiBox detector. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) Computer Vision-ECCV 2016, ECCV 2016, Lecture Notes in Computer Science(), vol. 9905. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46448-0_2.
16. Wojke, N., Bewley, A., Paulus, D.: Simple online and realtime tracking. In: 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA, pp. 3464–3468 (2016). https://doi.org/10.1109/ICIP.2016.7533003
17. Wojke, N., Bewley, A., Paulus, D.: Simple online and realtime tracking with a deep association metric. In: 2017 IEEE International Conference on Image Processing (ICIP), Beijing, China, pp. 3645–3649 (2017). https://doi.org/10.1109/ICIP.2017.8296962
18. Pereira, R., Carvalho, G., Garrote, L., Nunes, U.J.: Sort and deep-SORT based multi-object tracking for mobile robotics: evaluation with new data association metrics. Appl. Sci. 12, 1319 (2022). https://doi.org/10.3390/app12031319

Issue