半监督学习方法在作物图像处理中的应用进展

doi:10.3778/j.issn.1002-8331.2410-0016

摘要/Abstract

摘要： 半监督学习通过结合少量标注数据和大量未标注数据，显著减少数据标注需求和成本，同时提高模型的泛化能力和性能，为深度学习和机器学习在资源受限环境中的发展提供了新的路线。随着半监督学习在图像处理领域取得突破性进展，以现代农业为调查对象，探讨半监督学习方法在农业生产中的作用。根据半监督基础定义和假设提出半监督概念模型，同时按照利用监督信号的差异将半监督学习分为伪标签和无监督正则化。从目标检测、图像分割、图像分类和多任务联合学习出发，对农业作物管理、杂草管理、水果检测、植物健康、植物表型和叶片分类等具体任务进行全面回顾。分析并讨论了半监督学习在农业图像处理领域的未来研究方向。

关键词: 半监督, 作物, 图像处理, 目标检测, 图像分割, 图像分类

Abstract: Semi-supervised learning provides a new route for deep learning and machine learning in resource-constrained environments by combining a small amount of labeled data with a large amount of unlabeled data, significantly reducing the data labeling requirements and costs, while improving the generalization ability and performance of models. With the breakthrough progress of semi-supervised learning in the field of image processing, this review investigates the role of semi-supervised learning methods in agricultural production with modern agriculture as the object of investigation. A semi-supervised conceptual model is proposed, grounded in foundational definitions and assumptions. Semi-supervised learning is then classified into pseudo-labeling and unsupervised regularization, based on differences in the use of supervised signals. Starting from object detection, image segmentation, image classification, and multi-task joint learning, a comprehensive review is conducted on specific tasks such as agricultural crop management, weed management, fruit detection, plant health, plant phenotyping, and leaf classification. Future research directions for semi-supervised learning in agricultural image processing are analyzed and discussed.

Key words: semi-supervised, crop, , image processing, , target detection, , image segmentation, , image classification

陈曦, 刘建平, 周国民, 王健, 张越, 邢嘉璐. 半监督学习方法在作物图像处理中的应用进展[J]. 计算机工程与应用, 2025, 61(12): 45-68.

CHEN Xi, LIU Jianping, ZHOU Guomin, WANG Jian, ZHANG Yue, XING Jialu. Survey of Advances in Semi-Supervised Learning Methods for Crop Image Processing Applications[J]. Computer Engineering and Applications, 2025, 61(12): 45-68.

参考文献

[1] LI L, ZHANG Q, HUANG D F. A review of imaging techniques for plant phenotyping[J]. Sensors, 2014, 14(11): 20078-20111.
[2] KAMILARIS A, PRENAFETA-BOLDú F X. Deep learning in agriculture: a survey[J]. Computers and Electronics in Agriculture, 2018, 147: 70-90.
[3] SINGH A, GANAPATHYSUBRAMANIAN B, SINGH A K, et al. Machine learning for high-throughput stress phenotyping in plants[J]. Trends in Plant Science, 2016, 21(2): 110-124.
[4] JIANG Y, LI C Y, PATERSON A H. High throughput phenotyping of cotton plant height using depth images under field conditions[J]. Computers and Electronics in Agriculture, 2016, 130: 57-68.
[5] SLADOJEVIC S, ARSENOVIC M, ANDERLA A, et al. Deep neural networks based recognition of plant diseases by leaf image classification[J]. Computational Intelligence and Neuroscience, 2016, 2016(1): 3289801.
[6] MOHANTY S P, HUGHES D P, SALATHé M. Using deep learning for image-based plant disease detection[J]. Frontiers in Plant Science, 2016, 7: 1419.
[7] OUALI Y, HUDELOT C, TAMI M. An overview of deep semi-supervised learning[J]. arXiv:2006.05278, 2020.
[8] 葛静雯. 基于交叉监督的苹果树叶病害图像分类算法研究[D]. 长春: 吉林大学, 2023.
GE J W. Research on image classification algorithm of apple leaf diseases based on cross supervision[D]. Changchun: Jilin University, 2023.
[9] 王聃, 柴秀娟. 机器学习在植物病害识别研究中的应用[J]. 中国农机化学报, 2019, 40(9): 171-180.
WANG D, CHAI X J. Application of machine learning in plant diseases recognition[J]. Journal of Chinese Agricultural Mechanization, 2019, 40(9): 171-180.
[10] 陈冬梅, 林佳, 王海亮, 等. 基于多尺度数据集的虫害检测模型[J]. 农业工程学报, 2024, 40(5): 196-206.
CHEN D M, LIN J, WANG H L, et al. Pest detection model based on multi-scale dataset[J]. Transactions of the Chinese Society of Agricultural Engineering, 2024, 40(5): 196-206.
[11] LIU J P, WANG C Y, XING J L. YOLOv5-ACS: improved model for apple detection and positioning in apple forests in complex scenes[J]. Forests, 2023, 14(12): 2304.
[12] 贾少鹏, 高红菊, 杭潇. 基于深度学习的农作物病虫害图像识别技术研究进展[J]. 农业机械学报, 2019, 50(S1): 313-317.
JIA S P, GAO H J, HANG X. Research progress on image recognition technology of crop pests and diseases based on deep learning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(S1): 313-317.
[13] 慕君林, 马博, 王云飞, 等. 基于深度学习的农作物病虫害检测算法综述[J]. 农业机械学报, 2023, 54(S2): 301-313.
MU J L, MA B, WANG Y F, et al. Summary of detection algorithms of crop diseases and insect pests based on deep learning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(S2): 301-313.
[14] JIAO R S, ZHANG Y C, DING L, et al. Learning with limited annotations: a survey on deep semi-supervised learning for medical image segmentation[J]. Computers in Biology and Medicine, 2024, 169: 107840.
[15] DUARTE J M, BERTON L. A review of semi-supervised learning for text classification[J]. Artificial Intelligence Review, 2023: 1-69.
[16] CHAPELLE O, SCHOLKOPF B, ZIEN E. Semi-supervised learning (chapelle, O. et al., eds.; 2006) [book reviews][J]. IEEE Transactions on Neural Networks, 2009, 20(3): 542.
[17] OLIVIER C, BERNHARD S, ALEXANDER Z. Introduction to semi-supervised learning[M]//Semi-supervised learning. [S.l.]: The MIT Press, 2006: 1-12.
[18] YANG X L, SONG Z X, KING I, et al. A survey on deep semi-supervised learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2023, 35(9): 8934-8954.
[19] CUBUK E D, ZOPH B, SHLENS J, et al. Randaugment: practical automated data augmentation with a reduced search space[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway: IEEE, 2020: 3008-3017.
[20] HINTON G, VINYALS O, DEAN J. Distilling the knowledge in a neural network[J]. arXiv:1503.02531, 2015.
[21] TARVAINEN A, VALPOLA H. Mean teachers are better role models: weight-averaged consistency targets improve semi-supervised deep learning results[C]//Advances in Neural Information Processing Systems, 2017.
[22] LEO J, KALITA J. Incremental deep neural network learning using classification confidence thresholding[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022, 33(12): 7706-7716.
[23] FORT S, HU H Y, LAKSHMINARAYANAN B. Deep ensembles: a loss landscape perspective[J]. arXiv:1912. 02757, 2019.
[24] GAL Y, GHAHRAMANI Z. Dropout as a Bayesian approximation: representing model uncertainty in deep learning[C]//Proceedings of the 33rd International Conference on Machine Learning (ICML 2016), 2016: 1651-1660.
[25] 郑旭康, 李志忠, 秦俊豪. 基于半监督学习的梨叶病害检测[J]. 江苏农业科学, 2024, 52(5): 192-201.
ZHENG X K, LI Z Z , QIN J H. Study on pear leaf disease detection based on semi-supervised learning[J]. Jiangsu Agricultural Sciences, 2024, 52(5): 192-201.
[26] HAN K, LIU L, SONG Y Q, et al. An effective semi-supervised approach for liver CT image segmentation[J]. IEEE Journal of Biomedical and Health Informatics, 2022, 26(8): 3999-4007.
[27] 安瑞钰, 郝志斌. 基于半监督主动学习的小麦叶片病虫害区域分割方法[J]. 天津农学院学报, 2024, 31(2): 87-93.
AN R Y, HAO Z B. Regional segmentation method for wheat leaf diseases and insect pests based on semi-supervised active learning[J]. Journal of Tianjin Agricultural University, 2024, 31(2): 87-93.
[28] HUANG W, CHEN C, XIONG Z W, et al. Semi-supervised neuron segmentation via reinforced consistency learning[J]. IEEE Transactions on Medical Imaging, 2022, 41(11): 3016-3028.
[29] MAI X C, ZHU M L, YUAN Y X. CMCNet: colorization-aware mix-uncertainty-adaptive consistency network for semi-supervised fruit counting[J]. IEEE Transactions on Automation Science and Engineering, 2024, 21(4): 5766-5778.
[30] LIU T, ZHAI D L, HE F Y, et al. Semi-supervised learning methods for weed detection in turf[J]. Pest Management Science, 2024, 80(6): 2552-2562.
[31] BLUM A, MITCHELL T. Combining labeled and unlabeled data with co-training[C]//Proceedings of the Eleventh Annual Conference on Computational Learning Theory. New York: ACM, 1998: 92-100.
[32] LI D F, LI B L, FENG H Q, et al. Labour-saving detection of hybrid rice rows at the pollination stage based on a multi-perturbed semi-supervised model[J]. Computers and Electronics in Agriculture, 2023, 211: 107942.
[33] ZHANG Y Z, YANG L, CHEN J X, et al. Deep adversarial networks for biomedical image segmentation utilizing unannotated images[C]//Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2017). Cham: Springer, 2017: 408-416.
[34] MADSEN S L, MORTENSEN A K, J?RGENSEN R, et al. Disentangling information in artificial images of plant seedlings using semi-supervised GAN[J]. Remote Sensing, 2019, 11(22): 2671.
[35] GRANDVALET Y, BENGIO Y. Semi-supervised learning by entropy minimization[C]//Advances in Neural Information Processing Systems, 2004.
[36] JANG S Y, MOON G Y, KIM J O. Enhancing plant and disease segmentation through semi-supervised learning with feature distillation[C]//Proceedings of the 2023 IEEE International Conference on Consumer Electronics-Asia. Piscataway: IEEE, 2023: 1-3.
[37] ERON F, NOMAN M, DE OLIVEIRA R R, et al. Computer vision-aided intelligent monitoring of coffee: towards sustainable coffee production[J]. Scientia Horticulturae, 2024, 327: 112847.
[38] KHAKI S, PHAM H, HAN Y, et al. DeepCorn: a semi-supervised deep learning method for high-throughput image-based corn kernel counting and yield estimation[J]. Knowledge-Based Systems, 2021, 218: 106874.
[39] FOURATI F, MSEDDI W S, ATTIA R. Wheat head detection using deep, semi-supervised and ensemble learning[J]. Canadian Journal of Remote Sensing, 2021, 47(2): 198-208.
[40] CAI E Y, GUO J Q, YANG C Y, et al. Semi-supervised object detection for sorghum panicles in UAV imagery[C]//Proceedings of the 2023 IEEE International Geoscience and Remote Sensing Symposium(IGARSS 2023). Piscataway: IEEE, 2023: 6482-6485.
[41] XU X M, WANG L, LIANG X W, et al. Maize seedling leave counting based on semi-supervised learning and UAV RGB images[J]. Sustainability, 2023, 15(12): 9583.
[42] HU C S, THOMASSON J A, BAGAVATHIANNAN M V. A powerful image synthesis and semi-supervised learning pipeline for site-specific weed detection[J]. Computers and Electronics in Agriculture, 2021, 190: 106423.
[43] LIU T, JIN X J, ZHANG L Y, et al. Semi-supervised learning and attention mechanism for weed detection in wheat[J]. Crop Protection, 2023, 174: 106389.
[44] MENEZES G K, ASTOLFI G, MARTINS J A C, et al. Pseudo-label semi-supervised learning for soybean monitoring[J]. Smart Agricultural Technology, 2023, 4: 100216.
[45] SHOREWALA S, ASHFAQUE A, SIDHARTH R, et al. Weed density and distribution estimation for precision agriculture using semi-supervised learning[J]. IEEE Access, 2021, 9: 27971-27986.
[46] SUN T, XUE C Q, CHEN Y, et al. Cost-effective identification of the field maturity of tobacco leaves based on deep semi-supervised active learning and smartphone photograph[J]. Computers and Electronics in Agriculture, 2023, 215: 108373.
[47] JIANG Y P, CHEN S F, BIAN B, et al. Discrimination of tomato maturity using hyperspectral imaging combined with graph?based semi?supervised method considering class probability information[J]. Food Analytical Methods, 2021, 14(5): 968-983.
[48] ZHANG M M, XUE Y A, ZHAN Y Y, et al. Semi-supervised semantic segmentation-based remote sensing identification method for winter wheat planting area extraction[J]. Agronomy, 2023, 13(12): 2868.
[49] CAI Y L, ZENG F G, XIAO J Y, et al. Attention-aided semantic segmentation network for weed identification in pineapple field[J]. Computers and Electronics in Agriculture, 2023, 210: 107881.
[50] 蔡雨霖, 肖佳仪, 余超然, 等. 基于UANP-MT的半监督菜心杂草分割方法[J]. 农业工程学报, 2023, 39(11): 183-191.
CAI Y L, XIAO J Y, YU C R, et al. UANP-MT based semi-supervised image segmentation method for identifying weeds in cabbage field[J]. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(11): 183-191.
[51] CIARFUGLIA T A, MOTOI I M, SARACENI L, et al. Weakly and semi-supervised detection, segmentation and tracking of table grapes with limited and noisy data[J]. Computers and Electronics in Agriculture, 2023, 205: 107624.
[52] CASADO-GARCíA A, HERAS J, MILELLA A, et al. Semi-supervised deep learning and low-cost cameras for the semantic segmentation of natural images in viticulture[J]. Precision Agriculture, 2022, 23(6): 2001-2026.
[53] LI T, YANG J F, PENG X Q, et al. Prediction and early warning method for flea beetle based on semi-supervised learning algorithm[C]//Proceedings of the 2008 Fourth International Conference on Natural Computation. Piscataway: IEEE, 2008: 217-221.
[54] RUSTIA D J A, LU C Y, CHAO J J, et al. Online semi-supervised learning applied to an automated insect pest monitoring system[J]. Biosystems Engineering, 2021, 208: 28-44.
[55] QIN F, LIU D X, SUN B D, et al. Identification of alfalfa leaf diseases using image recognition technology[J]. PLoS One, 2016, 11(12): e0168274.
[56] ALAMOUDI S, HONG X, WEI H. Plant leaf recognition using texture features and semi-supervised spherical K-means clustering[C]//Proceedings of the 2020 International Joint Conference on Neural Networks. Piscataway: IEEE, 2020: 1-8.
[57] LI L L, GARIBALDI J M, HE D J. Leaf classification using multiple feature analysis based on semi-supervised clustering[J]. Journal of Intelligent & Fuzzy Systems, 29(4): 1465-1477.
[58] 吴惠思, 肖芳燕, 史周安, 等. 基于深度半监督学习的植物叶片自动识别[J]. 计算机辅助设计与图形学学报, 2023, 35(10): 1469-1478.
WU H S, XIAO F Y, SHI Z A, et al. Automatic leaf recognition based on deep semi-supervised learning[J]. Journal of Computer-Aided Design & Computer Graphics, 2023, 35(10): 1469-1478.
[59] BENCHALLAL F, HAFIANE A, RAGOT N, et al. ConvNeXt based semi-supervised approach with consistency regularization for weeds classification[J]. Expert Systems with Applications, 2024, 239: 122222.
[60] JIANG H H, ZHANG C Y, QIAO Y L, et al. CNN feature based graph convolutional network for weed and crop recognition in smart farming[J]. Computers and Electronics in Agriculture, 2020, 174: 105450.
[61] LOTTES P, STACHNISS C. Semi-supervised online visual crop and weed classification in precision farming exploiting plant arrangement[C]//Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2017: 5155-5161.
[62] JIANG K, YOU J, DORJ U O, et al. Detection of unknown strawberry diseases based on OpenMatch and two-head network for continual learning[J]. Frontiers in Plant Science, 2022, 13: 989086.
[63] 李洋, 侯文慧, 杨辉煌, 等. 密植环境下基于域自适应学习的番茄检测方法[J]. 农业工程学报, 2024, 40(13): 134-145.
LI Y, HOU W H, YANG H H, et al. Tomato detection method using domain adaptive learning for dense planting environments[J]. Transactions of the Chinese Society of Agricultural Engineering, 2024, 40(13): 134-145.
[64] ZHANG R, YAO M W, QIU Z J, et al. Wheat teacher: a one-stage anchor-based semi-supervised wheat head detector utilizing pseudo-labeling and consistency regularization methods[J]. Agriculture, 2024, 14(2): 327.
[65] KHANNA S, CHATTOPADHYAY C, KUNDU S. Enhancing fruit and vegetable detection in unconstrained environment with a novel dataset[J]. Scientia Horticulturae, 2024, 338: 113580.
[66] 张立才, 张欣, 陈孝玉龙, 等. 改进伪标签半监督方法的石榴生长阶段检测模型[J/OL]. 计算机工程与应用, 2024: 1-10 (2024-06-24). https://kns.cnki.net/kcms/detail/11.2127.tp.20240621.1803.016.html.
ZHANG L C, ZHANG X, CHEN X Y L, et al. Improved pseudo-label semi-supervised method for pomegranate growth stage detection modeling[J/OL]. Computer Engineering and Applications, 2024: 1-10 (2024-06-24). https://kns.cnki.net/kcms/detail/11.2127.tp.20240621.1803.016.html.
[67] ZHOU J L, HUANG H, SUN Y Q, et al. Mutual learning with memory for semi-supervised pest detection[J]. Frontiers in Plant Science, 2024, 15: 1369696.
[68] LI H Y, SHI F H. A DETR-like detector-based semi-supervised object detection method for Brassica Chinensis growth monitoring[J]. Computers and Electronics in Agriculture, 2024, 219: 108788.
[69] 江侯涛, 马善农. 基于半监督的植物病害智能检测研究[J]. 机电工程技术, 2024, 53(2): 221-224.
JIANG H T, MA S N. Research on intelligent detection of plant diseases based on semi-supervised learning[J]. Mechanical & Electrical Engineering Technology, 2024, 53(2): 221-224.
[70] GHANBARI A, SHIRDEL G H, MALEKI F. Semi-self-supervised domain adaptation: developing deep learning models with limited annotated data for wheat head segmentation[J]. Algorithms, 2024, 17(6): 267.
[71] 严露露, 朱赞彬, 冯世杰, 等. 基于改进FixMatch算法的半监督番茄病虫害识别[J/OL]. 江苏农业科学, 1-7[2024-12-01]. https://doi.org/10.15889/j.issn.1002-1302.2024.20.030.
YAN L L, ZHU Z B, FENG S J, et al. The semi-supervised tomato disease and pest recognition based on improved FixMatch algorithm[J/OL]. Jiangsu Agricultural Sciences, 1-7[2024-12-01]. https://doi.org/10.15889/j.issn.1002-1302. 2024.20.030.
[72] ILSEVER M, BAZ I. Consistency regularization based semi-supervised plant disease recognition[J]. Smart Agricultural Technology, 2024, 9: 100613.
[73] LI Y, CHAO X W. Semi-supervised few-shot learning approach for plant diseases recognition[J]. Plant Methods, 2021, 17(1): 68.
[74] CHEN D, LU Y Z, LI Z J, et al. Performance evaluation of deep transfer learning on multi-class identification of common weed species in cotton production systems[J]. Computers and Electronics in Agriculture, 2022, 198: 107091.
[75] ZHANG K X, LAMMERS K, CHU P Y, et al. Algorithm design and integration for a robotic apple harvesting system[C]//Proceedings of the 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2022: 9217-9224.
[76] ZHANG J Y, RAO Y, MAN C, et al. Identification of cucumber leaf diseases using deep learning and small sample size for agricultural Internet of Things[J]. International Journal of Distributed Sensor Networks, 2021, 17(4): 155014772110074.
[77] MINERVINI M, SCHARR H, TSAFTARIS S A. Image analysis: the new bottleneck in plant phenotyping [applications corner][J]. IEEE Signal Processing Magazine, 2015, 32(4): 126-131.
[78] KADIR A, NUGROHO L E, SUSANTO A, et al. Leaf classification using shape, color, and texture features[J]. arXiv:1401.4447, 2014.
[79] JOHANSON R, WILMS C, JOHANNSEN O, et al. S3AD: semi-supervised small apple detection in orchard environments[C]//Proceedings of the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision. Piscataway: IEEE, 2024: 7061-7070.
[80] SOHN K, ZHANG Z Z, LI C L, et al. A simple semi-supervised learning framework for object detection[J]. arXiv:2005.04757, 2020.
[81] WANG H, LYU S X, REN Y X. Paddy rice imagery dataset for panicle segmentation[J]. Agronomy, 2021, 11(8): 1542.
[82] MUKHTAR H, KHAN M Z, USMAN GHANI KHAN M, et al. Wheat plant counting using UAV images based on semi-supervised semantic segmentation[C]//Proceedings of the 2021 1st International Conference on Artificial Intelligence and Data Analytics. Piscataway: IEEE, 2021: 257-261.
[83] ZHANG S W, LEI Y K, ZHANG C L, et al. Semi-supervised orthogonal discriminant projection for plant leaf classification[J]. Pattern Analysis and Applications, 2016, 19(4): 953-961.
[84] ZHANG S W, LEI Y K, WU Y H. Semi-supervised locally discriminant projection for classification and recognition[J]. Knowledge-Based Systems, 2011, 24(2): 341-346.
[85] FENG Z Y, HUANG G H, CHI D C. Classification of the complex agricultural planting structure with a semi-supervised extreme learning machine framework[J]. Remote Sensing, 2020, 12(22): 3708.
[86] YANG J, CHEN Y. Tender leaf identification for early-spring green tea based on semi-supervised learning and image processing[J]. Agronomy, 2022, 12(8): 1958.
[87] LIU H L, ZHAN Y Z, XIA H F, et al. Self-supervised transformer-based pre-training method using latent semantic masking auto-encoder for pest and disease classification[J]. Computers and Electronics in Agriculture, 2022, 203: 107448.
[88] PATIL R R, KUMAR S. Rice-fusion: a multimodality data fusion framework for rice disease diagnosis[J]. IEEE Access, 2022, 10: 5207-5222.
[89] QING J J, DENG X L, LAN Y B, et al. GPT-aided diagnosis on agricultural image based on a new light YOLOPC[J]. Computers and Electronics in Agriculture, 2023, 213: 108168.