{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T16:26:45Z","timestamp":1787934405080,"version":"build-2784847793"},"reference-count":44,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T00:00:00Z","timestamp":1787875200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004543","name":"China Scholarship Council","doi-asserted-by":"publisher","award":["202408330448"],"award-info":[{"award-number":["202408330448"]}],"id":[{"id":"10.13039\/501100004543","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Scientific Research Start-up Project of Quzhou University","award":["BSYJ202347"],"award-info":[{"award-number":["BSYJ202347"]}]},{"name":"Project of Quzhou Municipal Bureau of Science and Technology","award":["2024K138"],"award-info":[{"award-number":["2024K138"]}]},{"name":"Project of Quzhou Municipal Bureau of Science and Technology","award":["2024K168"],"award-info":[{"award-number":["2024K168"]}]},{"name":"Zhejiang Provincial Natural Science Foundation of China","award":["LQN26F030045"],"award-info":[{"award-number":["LQN26F030045"]}]},{"name":"Zhejiang Provincial Natural Science Foundation of China","award":["LTGS24F020001"],"award-info":[{"award-number":["LTGS24F020001"]}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Microbiol."],"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To address the critical challenge of degraded detection performance caused by low-resolution microscopic inputs in field-based harmful algae monitoring, this paper proposes a cross-resolution knowledge distillation framework integrated with transfer learning. The method transfers fine-grained feature representations from a high-resolution YOLO teacher model to a low-resolution YOLO student model, thereby enabling accurate and efficient harmful algae identification at no additional inference cost.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Built upon the EMDS-7 environmental microorganism dataset, we employ a YOLO student network with a low input resolution, initialized with COCO pre-trained weights. A YOLO teacher model, trained at 1280\u202f\u00d7\u202f1,280 resolution, provides high-resolution feature supervision. To transfer the fine-grained spatial cues, we design a multi-scale attention-based distillation loss that aligns the spatial attention maps between student and teacher at the P3, P4, and P5 neck levels. The overall training objective is a weighted sum of the original detection loss and the feature distillation loss.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      On the test set of EMDS-7, the proposed method achieves a mAP50-95 of 0.78832 and a mAP50 of 0.87897, outperforming the undistilled baseline (0.78434 and 0.87341) as well as other methods such as YOLOv5s, YOLOv8s, YOLOv10s, YOLOv12s and YOLOv26s. Precision and recall reach 0.87542 and 0.84143, respectively. Ablation studies on the distillation weight\n                      <jats:inline-formula>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"M1\">\n                          <mml:mi>\u03bb<\/mml:mi>\n                        <\/mml:math>\n                      <\/jats:inline-formula>\n                      show that\n                      <jats:inline-formula>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"M2\">\n                          <mml:mi>\u03bb<\/mml:mi>\n                          <mml:mo>=<\/mml:mo>\n                          <mml:mn>0.5<\/mml:mn>\n                        <\/mml:math>\n                      <\/jats:inline-formula>\n                      gives the best performance, with lower or higher values leading to degraded results.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The proposed cross-resolution knowledge distillation framework effectively mitigates the fine-grained feature degradation inherent in low-resolution algae imagery, while transfer learning provides the student model with robust visual priors. The resulting lightweight detector, operating at low resolution, is well-suited for deployment on field-portable microscopic platforms. This work offers a viable pathway toward intelligent harmful algae monitoring in resource-constrained field settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmicb.2026.1914806","type":"journal-article","created":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T15:56:51Z","timestamp":1787932611000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Boosting low-resolution algae detection via cross-resolution knowledge distillation and transfer learning"],"prefix":"10.3389","volume":"17","author":[{"given":"Mingjie","family":"Jiang","sequence":"first","affiliation":[{"name":"College of Electrical and Information Engineering, Quzhou University","place":["Quzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mingxia","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Electrical and Information Engineering, Quzhou University","place":["Quzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuxun","family":"Wu","sequence":"additional","affiliation":[{"name":"College of Electrical and Information Engineering, Quzhou University","place":["Quzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haixia","family":"Zheng","sequence":"additional","affiliation":[{"name":"School of Computing, University of Portsmouth","place":["Portsmouth, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,28]]},"reference":[{"key":"ref1","first-page":"1","author":"An","year":"2025"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"2004.10934","DOI":"10.48550\/arXiv.2004.10934","article-title":"Yolov4: optimal speed and accuracy of object detection","author":"Bochkovskiy","year":"2020","journal-title":"arXiv preprint"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1663","DOI":"10.1111\/raq.12403","article-title":"Assessing risks and mitigating impacts of harmful algal blooms on mariculture and marine fisheries","volume":"12","author":"Brown","year":"2020","journal-title":"Rev. Aquac."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1007\/s11554-026-01910-3","article-title":"FFKD-net: lightweight small-object detection based on feature fusion and knowledge distillation","volume":"23","author":"Chen","year":"2026","journal-title":"J. Real-Time Image Process."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"329","DOI":"10.1021\/acsestwater.4c00853","article-title":"M-yolo v8s: classification and identification of different microalgae species based on the improved yolo v8s model for prevention of harmful algal blooms","volume":"5","author":"Dong","year":"2024","journal-title":"Acs Es&T Water"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"631","DOI":"10.1038\/s43017-024-00578-2","article-title":"Harmful algal blooms in inland waters","volume":"5","author":"Feng","year":"2024","journal-title":"Nat. Rev. Earth Environ."},{"key":"ref7","first-page":"1440","author":"Girshick","year":"2015"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"2508.03317","DOI":"10.48550\/arXiv.2508.03317","article-title":"Architectural insights into knowledge distillation for object detection: a comprehensive review","author":"Golizadeh","year":"2025","journal-title":"arXiv preprint"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1821911","DOI":"10.3389\/fmicb.2026.1821911","article-title":"A benchmark for environmental microorganism object detection on EMDS-7","volume":"17","author":"Guo","year":"2026","journal-title":"Front. Microbiol."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1427","DOI":"10.1093\/icesjms\/fsz171","article-title":"Developing a microscopic image dataset in support of intelligent phytoplankton detection using deep learning","volume":"77","author":"Han","year":"2019","journal-title":"ICES J. Mar. Sci."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"2308.04268","DOI":"10.48550\/arXiv.2308.04268","article-title":"Teacher-student architecture for knowledge distillation: a survey","author":"Hu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2025.3559616","article-title":"CRKD-YOLO: cross-resolution knowledge distillation for low-resolution remote sensing image object detection","volume":"74","author":"Huang","year":"2025","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1109\/TCYB.2025.3613498","article-title":"Social informer: pedestrian trajectory prediction by informer with adaptive trajectory probability region optimization","volume":"56","author":"Jiang","year":"2026","journal-title":"IEEE Trans. Cybern."},{"key":"ref14","first-page":"3732","author":"Kang","year":"2025"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"2409.12900","DOI":"10.48550\/arXiv.2409.12900","article-title":"Recognition of harmful phytoplankton from microscopic images using deep learning","author":"Khaldi","year":"2024","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"2407.20892","DOI":"10.48550\/arXiv.2407.20892","article-title":"What is YOLOv5: a deep look into the internal features of the popular object detector","author":"Khanam","year":"","journal-title":"arXiv preprint"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"2410.17725","DOI":"10.48550\/arXiv.2410.17725","article-title":"Yolov11: An overview of the key architectural enhancements","author":"Khanam","year":"","journal-title":"arXiv preprint"},{"key":"ref18","first-page":"743","author":"Kumar","year":"2023"},{"key":"ref19","first-page":"1","author":"Li","year":"2024"},{"key":"ref20","first-page":"2980","author":"Lin","year":"2017"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1007\/s44196-026-01350-8","article-title":"Enhanced detection of small low-resolution targets on water surfaces via hybrid feature enhancement and adaptive IoU","volume":"19","author":"Liu","year":"2026","journal-title":"Int. J. Comput. Intell. Syst."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"115171","DOI":"10.1016\/j.engappai.2026.115171","article-title":"Recent advances of YOLO-based object detection in remote sensing imagery: a comprehensive review","volume":"179","author":"Liu","year":"2026","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"152913","DOI":"10.1016\/j.scitotenv.2022.152913","article-title":"A review of the current and emerging detection methods of marine harmful microalgae","volume":"815","author":"Liu","year":"2022","journal-title":"Sci. Total Environ."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"798","DOI":"10.1007\/s11390-024-4158-5","article-title":"Knowledge distillation via hierarchical matching for small object detection","volume":"39","author":"Ma","year":"2024","journal-title":"J. Comput. Sci. Technol."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1007\/s11042-026-21520-2","article-title":"RT-FCOSH: bridging accuracy and efficiency in low-resolution object detection for autonomous driving","volume":"85","author":"Mboutayeb","year":"2026","journal-title":"Multimed. Tools Appl."},{"key":"ref26","first-page":"132","author":"Py","year":"2016"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1804.02767","DOI":"10.48550\/arXiv.1804.02767","article-title":"Yolov3: An incremental improvement","author":"Redmon","year":"2018","journal-title":"arXiv preprint"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"110710","DOI":"10.1016\/j.compeleceng.2025.110710","article-title":"Multi-source feature map distillation for enhanced low-resolution object recognition","volume":"128","author":"Ren","year":"2025","journal-title":"Comput. Electr. Eng."},{"key":"ref29","first-page":"1412.6550","article-title":"FitNets: hints for thin deep nets","author":"Romero","year":"2015","journal-title":"arXiv preprint"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"2509.25164","DOI":"10.48550\/arXiv.2509.25164","article-title":"YOLO26: key architectural enhancements and performance benchmarking for real-time object detection","author":"Sapkota","year":"2025","journal-title":"arXiv preprint"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"292","DOI":"10.3390\/s26010292","article-title":"Knowledge distillation in object detection: a survey from CNN to transformer","volume":"26","author":"Shehzadi","year":"2026","journal-title":"Sensors"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"116616","DOI":"10.1016\/j.marpolbul.2024.116616","article-title":"Enhancing microalgae classification accuracy in marine ecosystems through convolutional neural networks and support vector machines","volume":"205","author":"Sonmez","year":"2024","journal-title":"Mar. Pollut. Bull."},{"key":"ref33","first-page":"8606","author":"Sun","year":"2026"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"126038","DOI":"10.1016\/j.watres.2026.126038","article-title":"Towards scalable deep learning for automated microscopy in harmful algal bloom monitoring: data-centric workflow and multi-region generalisation","volume":"301","author":"Taheriashtiani","year":"2026","journal-title":"Water Res."},{"key":"ref35","first-page":"10781","author":"Tan","year":"2020"},{"key":"ref36","first-page":"1","author":"Tian","year":"2025"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"78433","DOI":"10.52202\/085713-2627","article-title":"Yolov12: attention-centric real-time object detectors","volume":"38","author":"Tian","year":"2026","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref38","first-page":"1","author":"Varghese","year":"2024"},{"key":"ref39","first-page":"16520","author":"Wang","year":"2024"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"104558","DOI":"10.1016\/j.algal.2026.104558","article-title":"Bibliometric analysis of freshwater algal blooms (1907\u20132024): Progress in monitoring and management","volume":"94","author":"Wang","year":"2026","journal-title":"Algal Res."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"107984","DOI":"10.52202\/079017-3429","article-title":"Yolov10: real-time end-to-end object detection","volume":"37","author":"Wang","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref42","first-page":"3","author":"Woo","year":"2018"},{"key":"ref43","doi-asserted-by":"publisher","first-page":"101749","DOI":"10.1016\/j.dwt.2026.101749","article-title":"AlgaeFusion-YOLO: research on algae detection in water bodies based on semantic-augmented feature fusion","volume":"326","author":"Wu","year":"2026","journal-title":"Desalin. Water Treat."},{"key":"ref44","doi-asserted-by":"publisher","first-page":"1084312","DOI":"10.3389\/fmicb.2023.1084312","article-title":"EMDS-7: environmental microorganism image dataset seventh version for multiple object detection evaluation","volume":"14","author":"Yang","year":"2023","journal-title":"Front. Microbiol."}],"container-title":["Frontiers in Microbiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmicb.2026.1914806\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T15:56:54Z","timestamp":1787932614000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmicb.2026.1914806\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,28]]},"references-count":44,"alternative-id":["10.3389\/fmicb.2026.1914806"],"URL":"https:\/\/doi.org\/10.3389\/fmicb.2026.1914806","relation":{},"ISSN":["1664-302X"],"issn-type":[{"value":"1664-302X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,28]]},"article-number":"1914806"}}