面向水环境管理的浮游生物图像采集与识别技术研究进展

doi:10.11996/JG.j.2095-302X.2026050971

图学学报 ›› 2026, Vol. 47 ›› Issue (5): 971-983.DOI: 10.11996/JG.j.2095-302X.2026050971

面向水环境管理的浮游生物图像采集与识别技术研究进展

石鑫迪¹, 赵浩然², 王烁阳¹, 陈栋¹, 王续澎², 周炜², 冯颂³, 施雪卿¹, 魏朋浩¹, 赵飞¹()

¹ 青岛理工大学环境与市政工程学院，山东青岛 266520
² 青岛理工大学信息与控制工程学院，山东青岛 266520
³ 中国科学院海洋研究所，山东青岛 266000

收稿日期:2026-04-15 接受日期:2026-07-15 出版日期:2026-10-31 发布日期:2026-10-26
通讯作者:赵飞，E-mail：zhaofei@qut.edu.cn
基金资助:
青岛市自然科学基金(25-1-1-268-zyyd-jch)

Research progress on plankton image acquisition and recognition technologies for water environmental management

SHI Xindi¹, ZHAO Haoran², WANG Shuoyang¹, CHEN Dong¹, WANG Xupeng², ZHOU Wei², FENG Song³, SHI Xueqing¹, WEI Penghao¹, ZHAO Fei¹()

¹ School of Environmental and Municipal Engineering, Qingdao University of Technology, Qingdao Shandong 266520, China
² School of Information and Control Engineering, Qingdao University of Technology, Qingdao Shandong 266520, China
³ Institute of Oceanology of the Chinese Academy of Sciences, Qingdao Shandong 266000, China

Received:2026-04-15 Accepted:2026-07-15 Published:2026-10-31 Online:2026-10-26
Contact: ZHAO Fei，E-mail：zhaofei@qut.edu.cn
Supported by:
Natural Science Foundation of Qingdao City(25-1-1-268-zyyd-jch)

摘要/Abstract

摘要：

浮游生物是水生态系统的重要组成部分，其群落结构、丰度变化和优势种演替能够反映水体营养状态、生态扰动和环境质量变化，在水生态监测、有害藻华预警和水质评价中具有重要指示作用。传统人工镜检方法依赖专业经验，存在识别效率低、主观性强、连续监测能力不足等问题，难以满足水环境管理对实时化、精细化和自动化监测的需求。随着成像设备和图像识别算法的发展，浮游生物监测逐渐由人工取样和显微鉴定向自动化成像与智能识别转变。在此基础上，围绕图像获取、数据支撑、方法演进和场景应用等环节，系统梳理浮游生物图像采集设备、公开数据集及图像识别技术的发展现状，重点分析识别方法由人工特征提取和传统机器学习向深度学习发展的过程，以及相关技术在群落监测与生态响应分析、有害藻华预警和水质评价等领域的应用进展。同时，显微成像、流式成像和原位成像等设备为浮游生物图像获取提供了不同技术路径，其中原位成像更适用于真实水体连续监测；经过专家标注或人工校核的公开数据集可为模型训练与性能评价提供较可靠的数据基础。深度学习已成为浮游生物图像识别的重要技术路线，在提高识别精度和自动化水平方面表现突出，但仍面临数据集标准化不足、模型泛化能力有限、现场部署成本较高及多源信息融合不足等问题。未来，自监督学习、ViT和边缘智能等技术可在降低人工标注依赖、增强复杂图像特征表达和提升现场部署能力等方面发挥作用。推动原位成像监测、自动识别方法与水质参数分析相结合，有助于进一步提升浮游生物图像识别技术在水环境监测、预警和评价中的应用价值。

关键词: 浮游生物, 图像识别, 机器学习, 深度学习, 水环境管理

Abstract:

Plankton are essential components of aquatic ecosystems, and variations in their community structure, abundance, and dominant species succession can reflect changes in water trophic status, ecological disturbances, and environmental quality. Therefore, plankton serve as important bioindicators in aquatic ecological monitoring, harmful algal bloom early warning, and water quality assessment. Traditional microscopic identification methods rely heavily on expert experience and are characterized by low efficiency, strong subjectivity, and limited capability for continuous monitoring, making it difficult to meet the it difficult real-time, fine-scale, and automated aquatic environmental management. As imaging technologies and image recognition algorithms advance, plankton monitoring has gradually shifted from manual sampling and microscopic identification toward automated imaging and intelligent recognition. Based on this development trend, this review systematically summarized the current progress of plankton imaging devices, publicly available datasets, and image recognition technologies from the perspectives of image acquisition, data support, methodological evolution, and application scenarios. The transition of recognition approaches from handcrafted feature extraction and traditional machine learning to deep learning-based methods was analyzed, along with recent applications in community monitoring and ecological response analysis, harmful algal bloom prediction, and water quality assessment. Overall, microscopic imaging, flow imaging, and in situ imaging technologies were shown to provide diverse approaches for plankton image acquisition, with in situ imaging being better suited to continuous monitoring in natural aquatic environments. Public datasets established through expert annotation or manual verification were found to provide reliable data foundations for model training and performance evaluation. Deep learning has become a major technical approach to plankton image recognition, demonstrating significant advantages in improving recognition accuracy and automation capability. However, challenges remain, including insufficient dataset standardization, limited model generalization ability, high field-deployment costs, and inadequate integration of multi-source information. In the future, emerging technologies such as self-supervised learning, Vision Transformer (ViT), and edge intelligence are expected to reduce dependence on manual annotation, enhance feature representation for complex images, and improve field deployment capabilities. Integrating in situ imaging monitoring, automated recognition methods, and water quality parameter analysis will further enhance the application potential of plankton image recognition technologies in aquatic environmental monitoring, early warning, and assessment.

Key words: plankton, image recognition, machine learning, deep learning, water environment management

中图分类号:

石鑫迪, 赵浩然, 王烁阳, 陈栋, 王续澎, 周炜, 冯颂, 施雪卿, 魏朋浩, 赵飞. 面向水环境管理的浮游生物图像采集与识别技术研究进展[J]. 图学学报, 2026, 47(5): 971-983.

SHI Xindi, ZHAO Haoran, WANG Shuoyang, CHEN Dong, WANG Xupeng, ZHOU Wei, FENG Song, SHI Xueqing, WEI Penghao, ZHAO Fei. Research progress on plankton image acquisition and recognition technologies for water environmental management[J]. Journal of Graphics, 2026, 47(5): 971-983.

图/表 9

参考文献 105

[1]	YI M L, LI L Q, LI H R, et al. Spatiotemporal variations of plankton communities in different water bodies of the Yellow River: structural characteristics, biogeographic patterns, environmental responses, and community assembly[J]. Journal of Hydrology, 2024, 640: 131702. DOI URL
[2]	GINCLEY B, KHAN F, HARTNETT E, et al. Introducing ARTiMiS: a low-cost flow imaging microscope for microalgal monitoring[J]. Environmental Science & Technology, 2024, 58(30): 13540-13551.
[3]	FU C P, LIU R S, FAN X, et al. Rethinking general underwater object detection: datasets, challenges, and solutions[J]. Neurocomputing, 2023, 517: 243-256. DOI URL
[4]	SOSA-TREJO D, BANDERA A, GONZÁLEZ M, et al. Vision-based techniques for automatic marine plankton classification[J]. Artificial Intelligence Review, 2023, 56(11): 12853-12884. DOI
[5]	CHENG K C, CHENG X M, HAO Q. A review of feature extraction technologies for plankton images[C]// 2018 International Conference on Information Hiding and Image Processing. New York: ACM, 2018: 48-56.
[6]	LI C, WANG K, XU N. A survey for the applications of content-based microscopic image analysis in microorganism classification domains[J]. Artificial Intelligence Review, 2019, 51(4): 577-646. DOI
[7]	EEROLA T, BATRAKHANOV D, VATANKHAH BARAZANDEH N, et al. Survey of automatic plankton image recognition: challenges, existing solutions and future perspectives[J]. Artificial Intelligence Review, 2024, 57(5): 114. DOI
[8]	YU Y, LI Y Z, SUN X, et al. MPT: a large-scale multiphytoplankton tracking benchmark[J]. Intelligent Marine Technology and Systems, 2024, 2(1): 35. DOI
[9]	COWEN R K, SPONAUGLE S, ROBINSON K L, et al. PlanktonSet 1.0: plankton imagery data collected from F.G. Walton Smith in Straits of Florida from 2014-06-03 to 2014-06-06 and used in the 2015 National Data Science Bowl[DB/OL]. (2015-04-28) [2026-01-14]. https://doi.org/10.7289/V5D21VJD.
[10]	ORENSTEIN E C, BEIJBOM O, PEACOCK E E, et al. WHOI-plankton- a large scale fine grained visual recognition benchmark dataset for plankton classification[EB/OL]. (2015-10-07) [2026-01-14]. https://arxiv.org/abs/1510.00745.
[11]	ELINEAU A, DESNOS C, JALABERT L, et al. ZooScanNet: plankton images captured with the ZooScan[DB/OL]. [2026-01-14]. https://agris.fao.org/search/en/records/647461cd425ec3c088ee7110.
[12]	LI Q, SUN X, DONG J Y, et al. Developing a microscopic image dataset in support of intelligent phytoplankton detection using deep learning[J]. ICES Journal of Marine Science, 2020, 77(4): 1427-1439. DOI URL
[13]	KIKO R, SCHRÖDER S M. UVP5 data sorted with EcoTaxa and MorphoCluster[DB/OL]. (2020-09-15) [2026-01-14]. https://doi.org/10.17882/73002.
[14]	KRAFT K, HARAGUCHI L, VELHONOJA O, et al. SYKE-plankton_IFCB_Utö_2021[DB/OL]. (2022-04-28) [2026-01-14]. https://doi.org/10.23728/B2SHARE.7C273B6F409C47E98A868D6517BE3AE3.
[15]	KRAFT K, VELHONOJA O, SEPPÄLÄ J, et al. SYKE- plankton_IFCB_2022[DB/OL]. (2022-02-01) [2026-01-14]. https://b2share.eudat.eu/records/xvnrp-7ga56.
[16]	LI J P, YANG Z Y, CHEN T. DYB-PlanktonNet[DB/OL]. (2021-04-08) [2026-01-14]. https://dx.doi.org/10.21227/875n-f104.
[17]	KYATHANAHALLY S P, HARDEMAN T, MERZ E, et al. Data for: deep learning classification of lake zooplankton[DB/OL]. (2021-11-03) [2026-01-14]. https://doi.org/10.25678/0004DY.
[18]	PLONUS R M, CONRADT J, HARMER A, et al. Automatic plankton image classification-can capsules and filters help coping with data set shift?[DB/OL]. (2021-01-14) [2026-01-14]. https://doi.org/10.5281/zenodo.4431509.
[19]	ZHENG H Y, WANG R C, YU Z B, et al. Automatic plankton image classification combining multiple view features via multiple kernel learning[J]. BMC Bioinformatics, 2017, 18(S16): 570. DOI URL
[20]	BUENO G, DENIZ O, PEDRAZA A, et al. Automated diatom classification (Part A): handcrafted feature approaches[J]. Applied Sciences, 2017, 7(8): 753. DOI URL
[21]	SÁNCHEZ C, CRISTÓBAL G, BUENO G. Diatom identification including life cycle stages through morphological and texture descriptors[J]. PeerJ, 2019, 7: e6770.
[22]	RAO A B, SRIHARI R K, ZHANG Z F. Spatial color histograms for content-based image retrieval[R]. Buffalo: State University of New York, CEDAR, 1999.
[23]	HU Q, DAVIS C. Automatic plankton image recognition with co-occurrence matrices and support vector machine[J]. Marine Ecology Progress Series, 2005, 295: 21-31. DOI URL
[24]	LI F, LIU D S, GUO X T, et al. Identification and visualization of environmental microplastics by Raman imaging based on hyperspectral unmixing coupled machine learning[J]. Journal of Hazardous Materials, 2024, 465: 133336. DOI URL
[25]	HALABAKU E, BYTYÇI E. Overfitting in machine learning: a comparative analysis of decision trees and random forests[J]. Intelligent Automation & Soft Computing, 2024, 39(6): 987-1006.
[26]	CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3): 273-297. DOI
[27]	HAJIZADEH R, AGHAGOLZADEH A, EZOJI M. Mutual neighborhood and modified majority voting based KNN classifier for multi-categories classification[J]. Pattern Analysis and Applications, 2022, 25(4): 773-793. DOI
[28]	刘乐源, 陈格格, 吴炜, 等. 数据分类分级技术研究综述[J]. 计算机科学, 2025, 52(9): 195-211.
	LIU L Y, CHEN G G, WU W, et al. Survey of data classification and grading studies[J]. Computer Science, 2025, 52(9): 195-211 (in Chinese).
[29]	PAVITHRA C, SARADHA M, ANTLINE NISHA B. Optimizing mail sorting with naive Bayes classifier and enhanced feature extraction method[J]. SN Computer Science, 2024, 5(7): 914. DOI
[30]	RIVAS-VILLAR D, ROUCO J, CARBALLEIRA R, et al. Fully automatic detection and classification of phytoplankton specimens in digital microscopy images[J]. Computer Methods and Programs in Biomedicine, 2021, 200: 105923. DOI URL
[31]	RIVAS-VILLAR D, ROUCO J, PENEDO M G, et al. Automatic detection of freshwater phytoplankton specimens in conventional microscopy images[J]. Sensors, 2020, 20(22): 6704. DOI URL
[32]	TEUWEN J, MORIAKOV N. Convolutional neural networks[M]//ZHOU S K, RUECKERT D, FICHTINGER G. Handbook of Medical Image Computing and Computer Assisted Intervention. Amsterdam: Elsevier, 2020: 481-501.
[33]	LEE H, PARK M, KIM J. Plankton classification on imbalanced large scale database via convolutional neural networks with transfer learning[C]// 2016 IEEE International Conference on Image Processing. New York: IEEE Press, 2016: 3713-3717.
[34]	KLOSTER M, LANGENKÄMPER D, ZUROWIETZ M, et al. Deep learning-based diatom taxonomy on virtual slides[J]. Scientific Reports, 2020, 10(1): 14416. DOI PMID
[35]	HYUN J, SEONG H, KIM E. Universal pooling-a new pooling method for convolutional neural networks[J]. Expert Systems with Applications, 2021, 180: 115084. DOI URL
[36]	周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229-1251.
	ZHOU F Y, JIN L P, DONG J. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6): 1229-1251 (in Chinese).
[37]	LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444. DOI
[38]	NANNI L, LUMINI A, BARCELLONA L, et al. Convolutional neural networks and vision transformers for plankton classification[J]. Ecological Informatics, 2025, 90: 103272. DOI URL
[39]	MA P L, LI C, RAHAMAN M M, et al. A state-of-the-art survey of object detection techniques in microorganism image analysis: from classical methods to deep learning approaches[J]. Artificial Intelligence Review, 2023, 56(2): 1627-1698. DOI
[40]	HIRATA N S T, FERNANDEZ M A, LOPES R M. Plankton image classification based on multiple segmentations[C]//2016 ICPR 2nd Workshop on Computer Vision for Analysis of Underwater Imagery. New York: IEEE Press, 2016: 55-60.
[41]	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]// The 26th International Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2012: 1097-1105.
[42]	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]// The 3rd International Conference on Learning Representations. San Diego: Computational and Biological Learning Society, 2015: 1-14.
[43]	SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2015: 1-9.
[44]	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2016: 770-778.
[45]	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2016: 779-788.
[46]	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2014: 580-587.
[47]	CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]// The 16th European Conference on Computer Vision. Cham: Springer, 2020: 213-229.
[48]	BADRINARAYANAN V, KENDALL A, CIPOLLA R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495. DOI PMID
[49]	HE K M, GKIOXARI G, DOLLÁR P, et al. Mask R-CNN[C]// 2017 IEEE International Conference on Computer Vision. New York: IEEE Press, 2017: 2980-2988.
[50]	CIRANNI M, GJERGJI A, MARACANI A, et al. In-domain self-supervised learning for plankton image classification on a budget[C]//2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops. New York: IEEE Press, 2025: 1503-1512.
[51]	KAREINEN J, EEROLA T, KRAFT K, et al. Self-supervised pretraining for fine-grained plankton recognition[C]// 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. New York: IEEE Press, 2025: 2122-2132.
[52]	OLSON R J, SOSIK H M. A submersible imaging-in-flow instrument to analyze nano-and microplankton: imaging FlowCytobot[J]. Limnology and Oceanography: Methods, 2007, 5(6): 195-203. DOI URL
[53]	BARTON A D, IRWIN A J, FINKEL Z V, et al. Anthropogenic climate change drives shift and shuffle in North Atlantic phytoplankton communities[J]. Proceedings of the National Academy of Sciences of the United States of America, 2016, 113(11): 2964-2969. DOI PMID
[54]	CAMPBELL L, HENRICHS D W, OLSON R J, et al. Continuous automated imaging-in-flow cytometry for detection and early warning of Karenia brevis blooms in the Gulf of Mexico[J]. Environmental Science and Pollution Research, 2013, 20(10): 6896-6902. DOI URL
[55]	DING G G, SHI Y H, LIU Z Q, et al. FE-YOLO: an efficient deep learning model based on feature-enhanced YOLOv7 for microalgae identification and detection[J]. Biomimetics, 2025, 10(1): 62. DOI URL
[56]	PARK J, LEE H, PARK C Y, et al. Algal morphological identification in watersheds for drinking water supply using neural architecture search for convolutional neural network[J]. Water, 2019, 11(7): 1338. DOI URL
[57]	PEDRAZA A, BUENO G, DENIZ O, et al. Automated diatom classification (Part B): a deep learning approach[J]. Applied Sciences, 2017, 7(5): 460. DOI URL
[58]	GÖRÖCS Z, TAMAMITSU M, BIANCO V, et al. A deep learning-enabled portable imaging flow cytometer for cost-effective, high-throughput, and label-free analysis of natural water samples[J]. Light: Science & Applications, 2018, 7(1): 66.
[59]	LUO J Y, IRISSON J O, GRAHAM B, et al. Automated plankton image analysis using convolutional neural networks[J]. Limnology and Oceanography: Methods, 2018, 16(12): 814-827. DOI URL
[60]	GONZÁLEZ P, ÁLVAREZ E, DÍEZ J, et al. Validation methods for plankton image classification systems[J]. Limnology and Oceanography: Methods, 2017, 15(3): 221-237. DOI URL
[61]	COLAS F, TARDIVEL M, PERCHOC J, et al. The ZooCAM, a new in-flow imaging system for fast onboard counting, sizing and classification of fish eggs and metazooplankton[J]. Progress in Oceanography, 2018, 166: 54-65. DOI URL
[62]	DYOMIN V, DAVYDOVA A, POLOVTSEV I, et al. Underwater holographic sensor for plankton studies in situ including accompanying measurements[J]. Sensors, 2021, 21(14): 4863. DOI URL
[63]	COWEN R K, GUIGAND C M. In situ ichthyoplankton imaging system (ISIIS): system design and preliminary results[J]. Limnology and Oceanography: Methods, 2008, 6(2): 126-132. DOI URL
[64]	LLORET L, HEREDIA I, AGUILAR F, et al. Convolutional neural networks for phytoplankton identification and classification[J]. Biodiversity Information Science and Standards, 2018, 2: e25762.
[65]	GORSKY G, OHMAN M D, PICHERAL M, et al. Digital zooplankton image analysis using the ZooScan integrated system[J]. Journal of Plankton Research, 2010, 32(3): 285-303. DOI URL
[66]	YANG Z Y, LI J P, CHEN T, et al. Contrastive learning-based image retrieval for automatic recognition of in situ marine plankton images[J]. ICES Journal of Marine Science, 2022, 79(10): 2643-2655. DOI URL
[67]	LYSENKO A V, OZNOBIKHIN M S, KIREEV E A, et al. Identification of Baikal phytoplankton inferred from computer vision methods and machine learning[J]. Limnology and Freshwater Biology, 2021(3): 1143-1146.
[68]	CAMOYING M G, YÑIGUEZ A T. FlowCAM optimization: attaining good quality images for higher taxonomic classification resolution of natural phytoplankton samples[J]. Limnology and Oceanography: Methods, 2016, 14(5): 305-314. DOI URL
[69]	CONRADT J, BÖRNER G, LÓPEZ-URRUTIA Á et al. Automated plankton classification with a dynamic optimization and adaptation cycle[J]. Frontiers in Marine Science, 2022, 9: 868420. DOI URL
[70]	DECROP W, LAGAISSE R, MORTELMANS J, et al. Automated image classification workflow for phytoplankton monitoring[J]. Frontiers in Marine Science, 2025, 12: 1699781. DOI URL
[71]	HU Q, DAVIS C. Accurate automatic quantification of taxa-specific plankton abundance using dual classification with correction[J]. Marine Ecology Progress Series, 2006, 306: 51-61. DOI URL
[72]	MACNEIL L, MISSAN S, LUO J L, et al. Plankton classification with high-throughput submersible holographic microscopy and transfer learning[J]. BMC Ecology and Evolution, 2021, 21(1): 123. DOI
[73]	MASOUDI M, GIERING S L C, EFTEKHARI N, et al. Optimizing plankton image classification with metadata- enhanced representation learning[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17: 17117-17133. DOI URL
[74]	MERKLI S, REYES M, POMATI F. Laboratory application of the Aquascope approach of automated imaging and classification for long-term plankton monitoring[EB/OL]. (2024-02-24) [2026-01-14]. https://doi.org/10.1101/2024.02.23.581739.
[75]	PITOIS S G, BLACKWELL R E, CLOSE H, et al. RAPID: real-time automated plankton identification dashboard using Edge AI at sea[J]. Frontiers in Marine Science, 2025, 11: 1513463. DOI URL
[76]	XU X F, LUO W L, REN Z H, et al. Intelligent detection and recognition of marine plankton by digital holography and deep learning[J]. Sensors, 2025, 25(7): 2325. DOI URL
[77]	BLASCHKO M B, HOLNESS G, MATTAR M A, et al. Automatic in situ identification of plankton[C]//2005 Seventh IEEE Workshops on Applications of Computer Vision - Volume 1. New York: IEEE Press, 2005: 79-86.
[78]	HENRICHS D W, ANGLÈS S, GAONKAR C C, et al. Application of a convolutional neural network to improve automated early warning of harmful algal blooms[J]. Environmental Science and Pollution Research, 2021, 28(22): 28544-28555. DOI
[79]	YANG M Y, WANG W S, GAO Q, et al. Automatic identification of harmful algae based on multiple convolutional neural networks and transfer learning[J]. Environmental Science and Pollution Research, 2023, 30(6): 15311-15324. DOI
[80]	刘扬, 孔凡洲, 于仁成, 等. 基于FlowCam影像和YOLOv3深度学习模型的多纹膝沟藻检测[J]. 海洋科学, 2024, 48(2): 15-23.
	LIU Y, KONG F Z, YU R C, et al. Detection of Gonyaulax polygramma using FlowCam and YOLOv3 model[J]. Marine Sciences, 2024, 48(2): 15-23 (in Chinese).
[81]	KRAFT K, SEPPÄLÄ J, HÄLLFORS H, et al. First application of IFCB high-frequency imaging-in-flow cytometry to investigate bloom-forming filamentous cyanobacteria in the Baltic Sea[J]. Frontiers in Marine Science, 2021, 8: 594144. DOI URL
[82]	STRICKLER J, COSTELLO J. Automatic classification of field-collected dinoflagellates by artificial neural network[J]. Marine Ecology Progress Series, 1996, 139: 307-309. DOI URL
[83]	KHAN F, GINCLEY B, BUSCH A, et al. Integrating machine learning with flow-imaging microscopy for automated monitoring of algal blooms[J]. Environmental Science & Technology, 2025, 59(37): 19885-19898. DOI URL
[84]	POMATI F, JOKELA J, SIMONA M, et al. An automated platform for phytoplankton ecology and aquatic ecosystem monitoring[J]. Environmental Science & Technology, 2011, 45(22): 9658-9665. DOI URL
[85]	DONG J H, WANG J S, LIN H M, et al. M-YOLO v8s: classification and identification of different microalgae species based on the improved YOLO v8s model for prevention of harmful algal blooms[J]. ACS ES&T Water, 2025, 5(1): 329-340.
[86]	QIAO Q S, YUAN A R, MA Y Y, et al. An automated field imaging microscope (AFIM) for harmful algal bloom (HAB) monitoring and aquaculture management[J]. Marine Pollution Bulletin, 2025, 217: 118104. DOI URL
[87]	GUNDUZ H, GUNAL S. A lightweight convolutional neural network (CNN) model for diatom classification: DiatomNet[J]. PeerJ Computer Science, 2024, 10: e1970.
[88]	FENG F J, WANG S, ZHANG X Q, et al. Fine-grained identification of benthic diatom scanning electron microscopy images using a deep learning framework[J]. Journal of Marine Science and Engineering, 2025, 13(6): 1095. DOI URL
[89]	CHEN Z, DU M, YANG X D, et al. Deep-learning-based automated tracking and counting of living plankton in natural aquatic environments[J]. Environmental Science & Technology, 2023, 57(46): 18048-18057. DOI URL
[90]	LIBREROS J, BUENO G, TRUJILLO M, et al. Automated identification and classification of diatoms from water resources[C]// The 23rd Iberoamerican Congress on Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Cham: Springer, 2018: 496-503.
[91]	KAHRAMAN İ, KARAŞ İ R, TURAN M K. Real-time protozoa detection from microscopic imaging using YOLOv4 algorithm[J]. Applied Sciences, 2024, 14(2): 607. DOI URL
[92]	COLTELLI P, BARSANTI L, EVANGELISTA V, et al. Water monitoring: automated and real time identification and classification of algae using digital microscopy[J]. Environmental Science: Processes & Impacts, 2014, 16(11): 2656-2665.
[93]	CHANG Y J, CHANG H S, CHANG C H, et al. A novel machine learning approach for assessing water quality using Daphnia swimming behavior[J]. Discover Water, 2025, 5(1): 115. DOI
[94]	VIJAYALAKSHMI M, SASITHRADEVI A. A comprehensive review on deep learning architecture for pre-processing of underwater images[J]. SN Computer Science, 2024, 5(5): 472. DOI
[95]	WANG M J, ZHANG K K, WEI H, et al. Underwater image quality optimization: researches, challenges, and future trends[J]. Image and Vision Computing, 2024, 146: 104995. DOI URL
[96]	MARACANI A, PASTORE V P, NATALE L, et al. In-domain versus out-of-domain transfer learning in plankton image classification[J]. Scientific Reports, 2023, 13(1): 10443. DOI PMID
[97]	宋巍, 邢晶晶, 杜艳玲, 等. 基于预处理图像惩罚的生成对抗网络水下图像增强[J]. 激光与光电子学进展, 2021, 58(12): 121002.
	SONG W, XING J J, DU Y L, et al. Underwater image enhancement based on generative adversarial network with preprocessed image penalty[J]. Laser & Optoelectronics Progress, 2021, 58(12): 121002 (in Chinese).
[98]	BOCHINSKI E, BACHA G, EISELEIN V, et al. Deep active learning for in situ plankton classification[C]// ICPR 2018 International Workshops on Pattern Recognition and Information Forensics. Cham: Springer, 2018: 5-15.
[99]	GUO C F, WEI B, YU K. Deep transfer learning for biology cross-domain image classification[J]. Journal of Control Science and Engineering, 2021, 2021(1): 2518837.
[100]	SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM: visual explanations from deep networks via gradient-based localization[C]// 2017 IEEE International Conference on Computer Vision. New York: IEEE Press, 2017: 618-626.
[101]	CAI S J, ZUO W M, ZHANG L. Higher-order integration of hierarchical convolutional activations for fine-grained visual categorization[C]// 2017 IEEE International Conference on Computer Vision. New York: IEEE Press, 2017: 511-520.
[102]	WANG N, CHEN T K, LIU S M, et al. Deep learning-based visual detection of marine organisms: a survey[J]. Neurocomputing, 2023, 532: 1-32. DOI URL
[103]	ZHENG J H, ZHAO R L, YANG G L, et al. An underwater image restoration deep learning network combining attention mechanism and brightness adjustment[J]. Journal of Marine Science and Engineering, 2024, 12(1): 7. DOI URL
[104]	PIERELLA KARLUSICH J J, LOMBARD F, IRISSON J O, et al. Coupling imaging and omics in plankton surveys: state-of-the-art, challenges, and future directions[J]. Frontiers in Marine Science, 2022, 9: 878803. DOI URL
[105]	HUO Y B, LV Q X, DONG J Y, et al. Enhancing phytoplankton recognition through a hybrid dataset and morphological description-driven prompt learning[J]. Journal of Marine Science and Engineering, 2025, 13(9): 1680. DOI URL

名称	工作场景	粒径范围	图像类型	图像采集后的识别方式
明场显微镜	实验室	300 nm~10 mm	单色/彩色	人工识别
Imaging FlowCytobot（IFCB）	原位/实验室	10~150 μm	单色	可结合识别模型自动分类
CytoSense/Cytobuoy	原位/实验室	1~800 μm	单色	配套软件辅助识别
FlowCam-Nano	实验室	300 nm~30 μm	单色	配套软件辅助识别
FlowCam	实验室	3~1 000 μm	单色/彩色	配套软件辅助识别
FlowCam-Macro	实验室	300 μm~5 mm	单色/彩色	配套软件辅助识别
ZooScan	实验室	150 μm~100 mm	单色	软件和模型识别分类
LISST-Holo2	原位	25~2 500 μm	全息图	全息图重建后再识别
UVP-5	原位	60 μm~20 mm	单色	采集后由识别模型分类
ISIIS	原位	60 μm~130 mm	单色	采集后由识别模型分类
CPICS	原位	30 μm~20 mm	彩色	采集后由识别模型分类
VPR	原位	30 μm~50 mm	单色视频	采集后由识别模型分类
LOKI	原位	50 μm~20 mm	单色	采集后由识别模型分类
普通光学显微镜+数码相机	实验	0.5 μm~1.0 mm	单色/彩色	人工识别或采集后模型分类
数字全息显微镜	原位/实验室	1~100 μm	全息图	全息图重建后再识别
全息-流式组合	原位/在线监测	0.5~50.0 μm	3D全息流速图	全息图重建后结合模型识别

名称	工作场景	粒径范围	图像类型	图像采集后的识别方式
明场显微镜	实验室	300 nm~10 mm	单色/彩色	人工识别
Imaging FlowCytobot（IFCB）	原位/实验室	10~150 μm	单色	可结合识别模型自动分类
CytoSense/Cytobuoy	原位/实验室	1~800 μm	单色	配套软件辅助识别
FlowCam-Nano	实验室	300 nm~30 μm	单色	配套软件辅助识别
FlowCam	实验室	3~1 000 μm	单色/彩色	配套软件辅助识别
FlowCam-Macro	实验室	300 μm~5 mm	单色/彩色	配套软件辅助识别
ZooScan	实验室	150 μm~100 mm	单色	软件和模型识别分类
LISST-Holo2	原位	25~2 500 μm	全息图	全息图重建后再识别
UVP-5	原位	60 μm~20 mm	单色	采集后由识别模型分类
ISIIS	原位	60 μm~130 mm	单色	采集后由识别模型分类
CPICS	原位	30 μm~20 mm	彩色	采集后由识别模型分类
VPR	原位	30 μm~50 mm	单色视频	采集后由识别模型分类
LOKI	原位	50 μm~20 mm	单色	采集后由识别模型分类
普通光学显微镜+数码相机	实验	0.5 μm~1.0 mm	单色/彩色	人工识别或采集后模型分类
数字全息显微镜	原位/实验室	1~100 μm	全息图	全息图重建后再识别
全息-流式组合	原位/在线监测	0.5~50.0 μm	3D全息流速图	全息图重建后结合模型识别

数据集名称	浮游生物类型	图像数量	类别数	成像平台	标注方式	采集区域	参考文献
Kaggle-Plankton	浮游动物	30 336	121	ISIIS-2	人工标注	美国佛罗里达海峡	[9]
WHOI-Plankton	浮游植物	3.5 M	103	IFCB	人工标注	美国马萨诸塞州	[10]
ZooScanNet	浮游动物	1.433 M	93	ZooScan	半自动+人工校核	法国	[11]
PMID2019	浮游植物	10 819	24	明场显微镜	人工标注	中国青岛胶州湾	[12]
UVP5/MC	浮游动物	1.588 M	65	UVP5	人工+自动标注混合	全球航次	[13]
SYKE-plankton_IFCB_Utö	浮游植物	57 000	50	IFCB	人工标注	芬兰、波罗的海	[14]
SYKE-plankton_IFCB	浮游植物	63 000	50	IFCB	人工标注	波罗的海	[15]
DYB-PlanktonNet	浮游动物	47 419	90	暗场显微镜	人工标注	中国大亚湾，南海	[16]
Lake-Zooplankton	浮游动物	17 943	35	双SPC	人工标注	瑞士格赖芬湖	[17]
Plonus	浮游动物	218 000	26	VPR	半自动+人工校核	北海-波罗的海	[18]

数据集名称	浮游生物类型	图像数量	类别数	成像平台	标注方式	采集区域	参考文献
Kaggle-Plankton	浮游动物	30 336	121	ISIIS-2	人工标注	美国佛罗里达海峡	[9]
WHOI-Plankton	浮游植物	3.5 M	103	IFCB	人工标注	美国马萨诸塞州	[10]
ZooScanNet	浮游动物	1.433 M	93	ZooScan	半自动+人工校核	法国	[11]
PMID2019	浮游植物	10 819	24	明场显微镜	人工标注	中国青岛胶州湾	[12]
UVP5/MC	浮游动物	1.588 M	65	UVP5	人工+自动标注混合	全球航次	[13]
SYKE-plankton_IFCB_Utö	浮游植物	57 000	50	IFCB	人工标注	芬兰、波罗的海	[14]
SYKE-plankton_IFCB	浮游植物	63 000	50	IFCB	人工标注	波罗的海	[15]
DYB-PlanktonNet	浮游动物	47 419	90	暗场显微镜	人工标注	中国大亚湾，南海	[16]
Lake-Zooplankton	浮游动物	17 943	35	双SPC	人工标注	瑞士格赖芬湖	[17]
Plonus	浮游动物	218 000	26	VPR	半自动+人工校核	北海-波罗的海	[18]

算法	支持输入的数据类型	优点	缺点	参考文献
SVM	数值型	稳定性强、精度高、适用性广	核函数和参数选择对结果影响大	[26]
RDF	数值型、分类型	抗拟合能力强、可输出特征重要性	计算成本高、预测速度较慢	[25]
KNN	数值型	无训练过程、实现简单	预测较慢，对k值敏感	[27]
DT	数值型、分类型	可解释性强、规则清晰	易过拟合、泛化弱、单棵树稳定性差	[28]
NB	数值型、计数型	计算快，对小规模数据表现良好	依赖特征独立假设、复杂特征下精度低	[29]

面向水环境管理的浮游生物图像采集与识别技术研究进展

Research progress on plankton image acquisition and recognition technologies for water environmental management

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 105

相关文章 15

编辑推荐

Metrics

本文评价

技术类别	输入的数据	典型任务	常用网络	参考文献
图像分类	单一目标的图像+图像级标注	物种筛查与鉴定、对水质快速评价	CNN（AlexNet，VGGNet，GoogLeNet，ResNet）	[41-44]
目标检测	多目标的原位图像+边界框标注	生物多样性快速评估、对目标物种计数与预警	YOLO系列、R-CNN、 DETR类	[45-47]
图像分割	多目标的原位图像+像素级标注	浮游生物形态功能群研究、生物量估算、健康状况评估	SegNet，Mask R-CNN	[48-49]

应用类型		应用场景与目的		成像平台/设备	技术方法	识别对象		参考文献
群落监测及生态响应分析		墨西哥湾海洋生态调查，实现快速、高分辨的浮游生物群落结构监测		ISIIS	SparseConvNet	浮游生物		[59]
群落监测及生态响应分析		比斯开湾与西班牙北部沿岸60个独立水样监测，估计未知水样8个类群的细胞丰度		FlowCam	SVM，RDF	8类浮游生物		[60]
		全球走航调查，获取浮游动物丰度、粒径谱，用于生态系统监测		ZooCAM	RDF+人工验证	中型浮游生物		[61]
		黑海6个站位海上监测试验，实现原位、非侵入、实时获取浮游生物丰度与生物量		数字全息相机	决策树	浮游生物单个体及链状体		[62]
		美国佛罗里达暖流核心区域，定量估算密度与微尺度分布		ISIIS	人工识别	中型浮游动物		[63]
		比利时北海观测站，替代传统人工显微镜识别，输出群落识别结果，提升监测效率与一致性		FlowCam	ResNet50	浮游植物		[64]
	法国Villefranche海湾监测，为浮游动物监测提供了一种高效、标准化和可重复的图像识别解决方案			ZooScan	随机森林+人工验证	33类浮游动物		[65]
	大亚湾与长江口海域监测，将浮游生物转化成实时、可验证和可更新的数据流			原位暗场成像系统	监督对比学习+ 图像检索	90类浮游生物		[66]
	俄罗斯贝加尔湖监测，构建淡水生态系统监测工具，支持水环境管理与保护决策			光学显微镜	Xception网络+ 迁移学习	浮游植物		[67]
	菲律宾两大沙丁鱼渔场，为后续实时浮游植物自动监测可复制、可扩展的技术模板			FlowCam	自定义的统计滤波器	硅藻、甲藻、蓝藻		[68]
	北海生态监测航次调查，提供了一种实用、可推广的浮游生物图像分类方案			FlowCam	CNN+DOC自适应机制	15类浮游生物		[69]
	比利时北海长期监测，替代人工镜检，实时输出细胞密度与群落结构			FlowCam	EfficientNetV2-B0卷积神经网络(CNN)	95类浮游植物		[70]
	美国乔治斯湾连续拖曳监测，实现自动、实时和高精度的浮游生物类群丰度估算			高倍拖曳相机(VPR)	双分类器融合+ 自动校正	6种浮游生物		[71]
	加拿大贝德福德湾监测，实现现场高通量、三维和非侵入式浮游生物分类基线			全息显微镜	CNN+迁移学习	19类浮游生物		[72]
	地中海、大西洋等多海域监测，实现高精度自动分类与计数，为大规模水环境自动监测提供解决方案			UVP6水下图像剖面仪	元数据增强自监督学习	23类浮游生物+碎屑		[73]
	瑞士多州湖泊监测，提供可溯源、可重分析和可共享的数字化结果，保障长期生态时间序列的延续性			Aquascope	Plankiformer CNN	58个浮游生物类群		[74]
	北海德国湾生态监测，为海洋生态监测与稀有物种识别提供了可扩展的自动化方案			VPR+TRIAXUS ROTV(遥控拖体)	CNN+概率过滤+Top-k策略	26类浮游生物及非生物颗粒		[18]
	对多种浮游生物图像数据集进行识别，构建通用浮游生物分类系统			公开浮游生物图像数据集	CNN与Vision Transformer集成模型	浮游生物		[38]
	海上走航实时监测，实现浮游动物图像自动分类与结果近实时传输，为海洋生态调查和自适应采样提供支持			Plankton Imager(Pi-10)	Edge AI+ResNet 分类模型	桡足、非侧足类浮游动物和碎屑类		[75]
	海洋浮游生物数字全息监测，实现浮游生物图像重建、目标检测与自动识别，为海洋生态健康监测提供支持			数字全息成像系统	A-Unet图像重建+YOLOv5目标检测	尾海鞘、箭虫、棘皮动物幼体等		[76]
	Woods Hole码头连续观测，用于研究物种演替、潮汐信号及有害藻华早期预警			IFCB	特征提取+SVM分类器训练	22类浮游植物		[52]
	构建浮游生物图像分类基准数据集，为自动识别模型训练与群落组成监测提供数据支撑			IFCB	人工特征方法与卷积神经网络分类模型	浮游生物		[10]
	分析气候变化驱动下浮游植物群落结构迁移与重组，为生态响应评估提供参考			长期观测资料与气候模型数据	人工整理与统计分析	北大西洋浮游植物群落		[53]
	海水原位自动监测，为海洋生态与健康评估提供高时空分辨率数据			FlowCam	特征提取+各种分类器训练	13类浮游生物		[77]
	瑞士格赖芬湖自动监测，实现高通量、标准化、低成本的浮游动物自动计数与分类			双倍数斯克里普斯浮游生物相机	EfficientNet系列CNN预训练+堆叠集成	35类淡水浮游生物		[17]
有害藻华预警	德州近岸赤潮藻监测，对其进行小时级浓度估计，实现赤潮早期预警			IFCB	CNN	有毒甲藻	[78]
	省市环境监测站、水厂，实现对有害藻类的精细识别，为水质监测的初步筛查提供支撑			实验室利用显微镜拍摄	5种CNN模型+ 迁移学习	11类有害藻、 31类无害藻	[79]
	山东半岛海域赤潮监测，验证图像识别技术对特定有害藻华进行早期监测			FlowCam	YOLOv3模型	有害藻种	[80]
	芬兰Utö岛自动观测站，对图像分类并提取像素体积换算生物量			IFCB	随机森林	蓝藻	[81]
	欧盟沿岸贝类养殖区例行监测，快速、低成本地自动判别有害藻华物种			光学显微镜+电视摄像机	特征提取+2种人工神经网络	23种有害甲藻	[82]
	北美五大湖3座饮用水厂取水口监测，为饮用水厂提供毒素产藻与堵塞藻的即时报警			流动成像显微镜(ARTiMiS)	CNN(5层CONV- MP-DO+2 Dense)	12类常见淡水藻	[83]
	西班牙多尼奥斯湖监测，高召回精准剔除非藻，为毒性藻监测提供单标本图像			普通光学显微镜	自适应阈值分割+Delaulay融合+RDF、SVM分类	51种淡水浮游植物及其干扰物	[31]
			瑞士卢加诺湖剖面监测，捕捉蓝藻水华与扰动事件，支撑湖泊管理与预警	Cytobuoy扫描流式细胞仪	PCA降维+K-means聚类+荧光阈值判定蓝藻	浮游植物个体以及蓝藻类群	[84]
			对不同微藻种类进行快速分类与识别，为有害藻华防控提供技术支持	光学显微镜	改进的YOLOv8s模型	不同的微藻种类	[85]
			香港海鱼养殖区现场监测，实现有害藻华物种自动识别，为水产养殖管理和赤潮预警提供支持	自动成像显微镜(AFIM)	Edge-AI-Chip分类器	有害藻华相关藻类	[86]
水质评价			土耳其河流水质生物监测，推断水体生态状态与水质等级，替代人工镜检	光学显微镜+相机拍摄	CNN(DiatomNet)	68类淡水硅藻	[87]
			杜罗河流域水质生物监测，快速评估水体生态状态与水质等级	光学显微镜+相机拍摄	Bagging决策树	80种淡水硅藻	[20]
			黄河流域（山东段）及其支流，识别硅藻种类，评估水质与生态系统健康状况	场发射扫描电子显微镜	自研KEGC网络架构	底栖硅藻	[88]
			河流污染监测，捕捉工业废水排放后种群数量，评价水质	自制微流控+高速相机	YOLOv5s6+DA检测	6类活体浮游生物	[89]
			实验室自动识别硅藻，将硅藻分类到属级，用于水质评估与生态监测	2种Nikon 显微镜	SCIRD-TS滤波+嵌套CNN	365张硅藻图像	[90]
			现场水质生物监测，实时检测生物并分类计数，提高监测效率	光学显微镜+相机拍摄	YOLOv4	8类淡水原生动物	[91]
			实验室环境监测模拟，为水质自动监测与公共供水安全提供了可直接部署的解决方案	显微镜+Basler彩色CCD	特征提取+SOM训练	23个藻类代表株	[92]
			利用水蚤运动行为视频分析评估水质污染水平，为低成本水质生物监测提供参考	手机视频/行为记录	运动特征提取+MLP分类模型	大型溞运动行为	[93]

[1]	吴雪, 刘子平. 基于改进的Deep-EIoU对多名足球运动员跟踪模型[J]. 图学学报, 2026, 47(5): 1001-1014.
[2]	杨志永, 刘迪, 姜杉, 周泽洋, 韩军, 朱珊, 陈超. 基于多椎体结构建模的CT-MRI脊柱多阶段配准方法[J]. 图学学报, 2026, 47(5): 1033-1042.
[3]	赵洁, 杨若荃, 武斌. 多特征融合的文档图像篡改区域分割方法[J]. 图学学报, 2026, 47(5): 984-993.
[4]	李煜华, 姜杉, 杨志永, 王禹泽, 周泽洋. 物理增强-深度协同的自由式三维超声重建[J]. 图学学报, 2026, 47(4): 726-735.
[5]	王雨涛, 杨超, 况立群, 杨晓文, 韩燮, 焦世超. 基于层次化对齐的三维模型零样本草图检索[J]. 图学学报, 2026, 47(4): 844-853.
[6]	李秀梅, 周正鑫, 孙军梅. 一种定位分支辅助的多任务协同图像伪造检测模型[J]. 图学学报, 2026, 47(3): 524-533.
[7]	吴文欢, 王文舒, 王舒鳌. 融合层次化双流注意力的单目深度估计方法[J]. 图学学报, 2026, 47(3): 553-563.
[8]	卢德辉, 宋琢, 黄志超, 田时雨, 李慧敏, 田茂, 邓逸川. 基于TrueSkill排序与深度学习的绿色工地主观视觉感知预测[J]. 图学学报, 2026, 47(3): 641-652.
[9]	王一芮, 华馨怡, 汤天宇, 王奕霖, 闫祉祺, 耿子涵, 陈星宇, 杨建明, 孙博文. 基于CiteSpace的国内外手术机器人交互设计分析比较研究[J]. 图学学报, 2026, 47(2): 251-263.
[10]	闫康, 曾理, 顾晓清. 基于跨域结构化深度字典学习的图像分类方法[J]. 图学学报, 2026, 47(2): 341-350.
[11]	庞敏, 李振堂, 张元, 崔晓康, 熊风光. 基于检索与变形技术的三维模型重构[J]. 图学学报, 2026, 47(2): 368-379.
[12]	董文益, 杨伟东, 唐冰慧, 王琦, 肖宏宇. 基于深度学习的肝脏局灶性病变检测方法综述[J]. 图学学报, 2026, 47(1): 1-16.
[13]	翟永杰, 王紫萱, 张祯琪, 周迅琪, 王乾铭. 融合双重注意力与加权动态卷积的车辆损伤分类模型[J]. 图学学报, 2026, 47(1): 17-28.
[14]	潘宇轩, 金锐, 刘雨, 张琳. 基于生成模型的无监督多视点立体视觉网络[J]. 图学学报, 2026, 47(1): 29-38.
[15]	酒明远, 吴国伟, 宋旭光, 李书攀, 徐明亮. 基于不确定性引导的智能强化主动学习图像分类方法[J]. 图学学报, 2026, 47(1): 47-56.