Hostname: page-component-f554764f5-nt87m Total loading time: 0 Render date: 2025-04-23T00:12:08.564Z Has data issue: false hasContentIssue false
  • English
  • Français

Knowledge distillation and student–teacher learning for weed detection in turf

Published online by Cambridge University Press:  29 October 2024

Danlan Zhai Open the ORCID record for Danlan Zhai [Opens in a new window] ,
Teng Liu ,
Feiyu He ,
Jinxu Wang ,
Xiaojun Jin  and
Jialin Yu
Danlan Zhai
Affiliation:
Intern, Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China
Teng Liu
Affiliation:
Research Assistant, Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China
Feiyu He
Affiliation:
Student, Department of Computer Science, Duke University, Durham, NC, USA
Jinxu Wang
Affiliation:
Research Assistant, Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China
Xiaojun Jin*
Affiliation:
Associate Professor, Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China
Jialin Yu*
Affiliation:
Professor and Principal Investigator, Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China
*
Corresponding authors: Xiaojun Jin; Email: [email protected]; Jialin Yu; Email: [email protected]
Corresponding authors: Xiaojun Jin; Email: [email protected]; Jialin Yu; Email: [email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

Machine vision–based herbicide applications relying on object detection or image classification deep convolutional neural networks (DCNNs) demand high memory and computational resources, resulting in lengthy inference times. To tackle these challenges, this study assessed the effectiveness of three teacher models, each trained on datasets of varying sizes, including D-20k (comprising 10,000 true-positive and true-negative images) and D-10k (comprising 5,000 true-positive and true-negative images). Additionally, knowledge distillation was performed on their corresponding student models across a range of temperature settings. After the process of student–teacher learning, the parameters of all student models were reduced. ResNet18 not only achieved higher accuracy (ACC ≥ 0.989) but also maintained higher frames per second (FPS ≥ 742.9) under its optimal temperature condition (T = 1). Overall, the results suggest that employing knowledge distillation in the machine vision models enabled accurate and reliable weed detection in turf while reducing the need for extensive computational resources, thereby facilitating real-time weed detection and contributing to the development of smart, machine vision–based sprayers.

Keywords

Deep learningneural networksturf managementprecision herbicide application
Type
Research Article
Information
Weed Science , Volume 72 , Issue 6 , November 2024 , pp. 804 - 815
Check for updates
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of Weed Science Society of America

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

Footnotes

*

These authors contributed equally to this work.

References

Al-Badri, AH, Ismail, NA, Al-Dulaimi, K, Salman, GA, Khan, A, Al-Sabaawi, A, Salam, MSH (2022) Classification of weed using machine learning techniques: a review—challenges, current and future potential techniques. J Plant Dis Prot 129:745–768 CrossRefGoogle Scholar
Alengebawy, A, Abdelkhalek, ST, Qureshi, SR, Wang, MQ (2021) Heavy metals and pesticides toxicity in agricultural soil and plants: ecological risks and human health implications. Toxics 9:42 CrossRefGoogle ScholarPubMed
Angarano, S, Martini, M, Navone, A, Chiaberge, M (2023) Domain generalization for crop segmentation with knowledge distillation. arXiv:2304.01029Google Scholar
Ba, J, Caruana, R (2014) Do deep nets really need to be deep? Pages 2654–2662 in Proceedings of the 28th Annual Conference on Neural Information Processing Systems. New York: Curran AssociatesGoogle Scholar
Bakhshipour, A, Jafari, A, Nassiri, SM, Zare, D (2017) Weed segmentation using texture features extracted from wavelet sub-images. Biosyst Eng 157:1–12 CrossRefGoogle Scholar
Brosnan, JT, Elmore, MT, Bagavathiannan, MV (2020) Herbicide-resistant weeds in turfgrass: current status and emerging threats. Weed Technol 34:424–430 CrossRefGoogle Scholar
Buciluǎ, C, Caruana, R, Niculescu-Mizil, A (2006) Model compression. Pages 535–541 in Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: Association for Computing MachineryCrossRefGoogle Scholar
Cai, X, Zhu, Y, Liu, S, Yu, Z, Xu, Y (2024) FastSegFormer: a knowledge distillation-based method for real-time semantic segmentation of surface defects in navel oranges. Comput Electron Agric 217:108604 CrossRefGoogle Scholar
Chen, J, Ran, X (2019) Deep learning with edge computing: a review. Proc IEEE 107:1655–1674 CrossRefGoogle Scholar
Cho, JH, Hariharan, B (2019) On the efficacy of knowledge distillation. Pages 4794–4802 in Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Washington, DC: IEEE Computer SocietyCrossRefGoogle Scholar
Dai, X, Xu, Y, Zheng, J, Song, H (2019) Analysis of the variability of pesticide concentration downstream of inline mixers for direct nozzle injection systems. Biosyst Eng 180:59–69 CrossRefGoogle Scholar
Deng, J, Dong, W, Socher, R, Li, LJ, Li, K, Li, FF (2009) Imagenet: a large-scale hierarchical image database. Pages 248–255 in Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: Institute of Electrical and Electronics EngineersCrossRefGoogle Scholar
El-Haggar, S, Samaha, A (2019) Sustainable urban community development guidelines. Pages 75–102 in Amer, M, ed. Roadmap for Global Sustainability—Rise of the Green Communities. Cham, Switzerland: Springer CrossRefGoogle Scholar
El-Rashidy, N, El-Sappagh, S, Islam, SR, El-Bakry, HM, Abdelrazek, S (2020) End-to-end deep learning framework for coronavirus (COVID-19) detection and monitoring. Electronics 9:1439 CrossRefGoogle Scholar
Ghofrani, A, Toroghi, RM (2022) Knowledge distillation in plant disease recognition. Neural Comput Appl 34:14287–14296 CrossRefGoogle Scholar
Hamuda, E, Glavin, M, Jones, E (2016) A survey of image processing techniques for plant extraction and segmentation in the field. Comput Electron Agric 125:184–199 CrossRefGoogle Scholar
Hasan, AM, Sohel, F, Diepeveen, D, Laga, H, Jones, MG (2021) A survey of deep learning techniques for weed detection from images. Comput Electron Agric 184:106067 CrossRefGoogle Scholar
Hasanuzzaman, M, Mohsin, SM, Bhuyan, MB, Bhuiyan, TF, Anee, TI, Masud, AAC, Nahar, K (2020) Phytotoxicity, environmental and health hazards of herbicides: challenges and ways forward. Pages 55–99 in Prasad, MNV, ed. Agrochemicals: Detection, Treatment and Remediation. Oxford, UK: Elsevier CrossRefGoogle Scholar
He, K, Zhang, X, Ren, S, Sun, J (2016) Deep residual learning for image recognition. Pages 770–778 in Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: Institute of Electrical and Electronics EngineersCrossRefGoogle Scholar
Hinton, G, Deng, L, Yu, D, Dahl, GE, Mohamed, AR, Jaitly, N, Senior, A, Vanhoucke, V, Nguyen, P, Sainath, TN (2012) Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Proc Mag 29:82–97 CrossRefGoogle Scholar
Hinton, G, Vinyals, O, Dean, J (2015) Distilling the knowledge in a neural network. arXiv:1503.02531 [stat.ML]Google Scholar
Huang, G, Liu, Z, Van Der Maaten, L, Weinberger, KQ (2017) Densely connected convolutional networks. Pages 4700–4708 in Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: Institute of Electrical and Electronics EngineersCrossRefGoogle Scholar
Jiang, H, Jiang, X, Ru, Y, Wang, J, Xu, L, Zhou, H (2020) Application of hyperspectral imaging for detecting and visualizing leaf lard adulteration in minced pork. Infrared Phys Technol 110:103467 CrossRefGoogle Scholar
Jin, X, Bagavathiannan, M, McCullough, PE, Chen, Y, Yu, J (2022a) A deep learning-based method for classification, detection, and localization of weeds in turfgrass. Pest Manag Sci 78:4809–4821 CrossRefGoogle ScholarPubMed
Jin, X, Han, K, Zhao, H, Wang, Y, Chen, Y, Yu, J (2024) Detection and coverage estimation of purple nutsedge in turf with image classification neural networks. Pest Manag Sci 80:3504–3515 CrossRefGoogle ScholarPubMed
Jin, X, Liu, T, Chen, Y, Yu, J (2022b) Deep learning-based weed detection in turf: a review. Agronomy 12:3051 CrossRefGoogle Scholar
Jin, X, Liu, T, Yang, Z, Xie, J, Bagavathiannan, M, Hong, X, Xu, Z, Chen, X, Yu, J, Chen, Y (2023) Precision weed control using a smart sprayer in dormant bermudagrass turf. Crop Prot 172:106302 CrossRefGoogle Scholar
Jung, JY, Lee, SH, Kim, JO (2022) Plant leaf segmentation using knowledge distillation. Pages 1–3 in Proceedings of the 2022 IEEE International Conference on Consumer Electronics—Asia. Piscataway, NJ: Institute of Electrical and Electronics EngineersCrossRefGoogle Scholar
Kakarla, SC, Costa, L, Ampatzidis, Y, Zhang, Z (2022) Applications of UAVs and machine learning in agriculture. Pages 1–19 in Zhang Z, Liu H, Yang C, Ampatzidis Y, Zhou J, Jiang Y, eds. Unmanned Aerial Systems in Precision Agriculture. Singapore: SpringerCrossRefGoogle Scholar
Khanam, R, Hussain, M, Hill, R, Allen, P (2024) A comprehensive review of convolutional neural networks for defect detection in industrial applications. IEEE Access 12:94250–94295 CrossRefGoogle Scholar
Krichen, M (2023) Convolutional neural networks: a survey. Computers 12:151 CrossRefGoogle Scholar
Li, Z, Li, X, Yang, L, Zhao, B, Song, R, Luo, L, Li, J, Yang, J (2023) Curriculum temperature for knowledge distillation. Pages 1504–1512 in Proceedings of the 37th AAAI Conference on Artificial Intelligence. Washington, DC: Association for the Advancement of Artificial Intelligence PressCrossRefGoogle Scholar
Liu, B, Bruch, R (2020) Weed detection for selective spraying: a review. Curr Robot Rep 1:19–26 CrossRefGoogle Scholar
McCullough, PE, Yu, J, Raymer, PL, Chen, Z (2016) First report of ACCase-resistant goosegrass (Eleusine indica) in the United States. Weed Sci 64:399–408 CrossRefGoogle Scholar
McCullough, PE, Yu, J, Shilling, DG, Czarnota, MA, Johnston, CR (2015) Biochemical effects of imazapic on bermudagrass growth regulation, broomsedge (Andropogon virginicus) control, and MSMA antagonism. Weed Sci 63:596–603 CrossRefGoogle Scholar
McElroy, J, Martins, D (2013) Use of herbicides on turfgrass. Planta Daninha 31:455–467 CrossRefGoogle Scholar
Mennan, H, Jabran, K, Zandstra, BH, Pala, F (2020) Non-chemical weed management in vegetables by using cover crops: a review. Agronomy 10:257 CrossRefGoogle Scholar
Monteiro, JA (2017) Ecosystem services from turfgrass landscapes. Urban For Urban Green 26:151–157 CrossRefGoogle Scholar
Nithya, R, Santhi, B, Manikandan, R, Rahimi, M, Gandomi, AH (2022) Computer vision system for mango fruit defect detection using deep convolutional neural network. Foods 11:3483 CrossRefGoogle ScholarPubMed
Pantazi, XE, Moshou, D, Bravo, C (2016) Active learning system for weed species recognition based on hyperspectral sensing. Biosyst Eng 146:193–202 CrossRefGoogle Scholar
Partel, V, Kakarla, SC, Ampatzidis, Y (2019) Development and evaluation of a low-cost and smart technology for precision weed management utilizing artificial intelligence. Comput Electron Agric 157:339–350 CrossRefGoogle Scholar
Perez, A, Lopez, F, Benlloch, J, Christensen, S (2000) Colour and shape analysis techniques for weed detection in cereal fields. Comput Electron Agric 25:197–212 CrossRefGoogle Scholar
Pincetl, S, Gillespie, TW, Pataki, DE, Porse, E, Jia, S, Kidera, E, Nobles, N, Rodriguez, J, Choi, DA (2019) Evaluating the effects of turf-replacement programs in Los Angeles. Landscape Urban Plan 185:210–221 CrossRefGoogle Scholar
Reed, TV, Yu, J, McCullough, PE (2013) Aminocyclopyrachlor efficacy for controlling Virginia buttonweed (Diodia virginiana) and smooth crabgrass (Digitaria ischaemum) in tall fescue. Weed Technol 27:488–491 CrossRefGoogle Scholar
Shakarami, A, Shahidinejad, A, Ghobaei-Arani, M (2021) An autonomous computation offloading strategy in mobile edge computing: a deep learning-based hybrid approach. J Netw Comput Appl 178:102974 CrossRefGoogle Scholar
Shuping, F, Yu, R, Chenming, H, Fengbo, Y (2023) Planning of takeoff/landing site location, dispatch route, and spraying route for a pesticide application helicopter. Eur J Agron 146:126814 CrossRefGoogle Scholar
Sokolova, M, Lapalme, G (2009) A systematic analysis of performance measures for classification tasks. Inform Process Manag 45:427–437 CrossRefGoogle Scholar
Stewart, R, Nowlan, A, Bacchus, P, Ducasse, Q, Komendantskaya, E (2021) Optimising hardware accelerated neural networks with quantisation and a knowledge distillation evolutionary algorithm. Electronics 10:396 CrossRefGoogle Scholar
Stier, JC, Steinke, K, Ervin, EH, Higginson, FR, McMaugh, PE (2013) Turfgrass benefits and issues. Pages 105–145 in Stier JC, Horgan BP, Bonos SA, eds. Turfgrass: Biology, Use, and Management. Wiley Online Books, https://doi.org/10.2134/agronmonogr56 CrossRefGoogle Scholar
Tan, M, Le, Q (2019) Efficientnet: rethinking model scaling for convolutional neural networks. Pages 6105–6114 in Proceedings of the 36th International Conference on Machine Learning. Maastricht, Netherlands: ML Research PressGoogle Scholar
Tang, JL, Chen, XQ, Miao, RH, Wang, D (2016) Weed detection using image processing under different illumination for site-specific areas spraying. Comput Electron Agric 122:103–111 CrossRefGoogle Scholar
Tate, TM, McCullough, PE, Harrison, ML, Chen, Z, Raymer, PL (2021) Characterization of mutations conferring inherent resistance to acetyl coenzyme A carboxylase-inhibiting herbicides in turfgrass and grassy weeds. Crop Sci 61:3164–3178 CrossRefGoogle Scholar
Tulbure, AA, Tulbure, AA, Dulf, EH (2022) A review on modern defect detection models using DCNNs–deep convolutional neural networks. J Adv Res 35:33–48 CrossRefGoogle ScholarPubMed
Upadhyay, A, Sunil, GC, Zhang, Y, Koparan, C, Sun, X (2024a) Development and evaluation of a machine vision and deep learning-based smart sprayer system for site-specific weed management in row crops: an edge computing approach. J Agric Food Res 18:101331 Google Scholar
Upadhyay, A, Zhang, Y, Koparan, C, Rai, N, Howatt, K, Bajwa, S, Sun, X (2024b) Advances in ground robotic technologies for site-specific weed management in precision agriculture: a review. Comput Electron Agric 225:109363 CrossRefGoogle Scholar
Urban, G, Geras, KJ, Kahou, SE, Aslan, O, Wang, S, Caruana, R, Mohamed, A, Philipose, M, Richardson, M (2017) Do deep convolutional nets really need to be deep and convolutional? arXiv:1603.05691 [stat.ML]Google Scholar
[USEPA] U.S. Environmental Protection Agency (2023) Ingredients Used in Pesticide Products—Atrazine. http://www.epa.gov/ingredients-used-pesticide-products/atrazine. Accessed: March 10, 2024Google Scholar
Wei, X, Zhang, H, Shi, C, Yang, X, Han, H, Li, B (2022) A lightweight flower classification model based on improved knowledge distillation. Pages 2236–2239 in Proceedings of the IEEE 10th Joint International Information Technology and Artificial Intelligence Conference. Piscataway, NJ: Institute of Electrical and Electronics EngineersCrossRefGoogle Scholar
Xie, S, Hu, C, Bagavathiannan, M, Song, D (2021) Toward robotic weed control: detection of nutsedge weed in bermudagrass turf using inaccurate and insufficient training data. IEEE Robot Autom Lett 6:7365–7372 CrossRefGoogle Scholar
Yang, G, Wang, B, Qiao, S, Qu, L, Han, N, Yuan, G, Li, H, Wu, T, Peng, Y (2022) Distilled and filtered deep neural networks for real-time object detection in edge computing. Neurocomputing 505:225–237 CrossRefGoogle Scholar
Yu, J, McCullough, PE (2016) Triclopyr reduces foliar bleaching from mesotrione and enhances efficacy for smooth crabgrass control by altering uptake and translocation. Weed Technol 30:516–523 CrossRefGoogle Scholar
Yu, J, McCullough, PE, Czarnota, MA (2018) Annual bluegrass (Poa annua) biotypes exhibit differential levels of susceptibility and biochemical responses to protoporphyrinogen oxidase inhibitors. Weed Sci 66:574–580 CrossRefGoogle Scholar
Yu, J, Schumann, AW, Cao, Z, Sharpe, SM, Boyd, NS (2019a) Weed detection in perennial ryegrass with deep learning convolutional neural network. Front Plant Sci 10:1422 CrossRefGoogle ScholarPubMed
Yu, J, Schumann, AW, Sharpe, SM, Li, X, Boyd, NS (2020) Detection of grassy weeds in bermudagrass with deep convolutional neural networks. Weed Sci 68:545–552 CrossRefGoogle Scholar
Yu, J, Sharpe, SM, Schumann, AW, Boyd, NS (2019b) Deep learning for image-based weed detection in turfgrass. Eur J Agron 104:78–84 CrossRefGoogle Scholar
Yu, J, Sharpe, SM, Schumann, AW, Boyd, NS (2019c) Detection of broadleaf weeds growing in turfgrass with convolutional neural networks. Pest Manag Sci 75:2211–2218 CrossRefGoogle ScholarPubMed
Zhou, W, Song, C, Song, K, Wen, N, Sun, X, Gao, P (2023) Surface defect detection system for carrot combine harvest based on multi-stage knowledge distillation. Foods 12:793 CrossRefGoogle ScholarPubMed