{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"Research Square"}],"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T17:14:09Z","timestamp":1776100449676,"version":"3.50.1"},"posted":{"date-parts":[[2025,9,19]]},"group-title":"In Review","reference-count":22,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2025,9,19]],"date-time":"2025-09-19T00:00:00Z","timestamp":1758240000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,9,12]]},"abstract":"<title>Abstract<\/title>\n                <p>Monitoring solar phenomena, such as sunspots and active regions, is crucial for ensuring astronaut safety, telecommunications reliability, and predicting terrestrial events like auroras. Traditional methods for detecting these phenomena have limitations in accuracy and baseline maintenance. This paper presents a novel deep learning object detection method that leverages multispectral image data from satellites to enhance the detection of \"sunspots\" and active regions. Utilizing images from the SDO satellite and annotations from the DeepSDO dataset, we constructed a new dataset composed of aligned observations from HMI Ic, AIA 211\\,\\AA, and AIA 335\\,\\AA. We adapted and developed a stock YOLOv5-based model capable of handling and fusing any number of input images. Two fusion methodologies, early and late fusion, and three different fusion modules --- CatFuse (simple concatenation), CBAMC (CBAM-based module), and TransEnc (transformer encoder) --- were implemented and tested. Our critical evaluation of the models, supported by statistical analysis, proved the developed models to be statistically significantly different among themselves at a p-value of 0.05, and helped us to identify the best-performing model: CatFuse with early fusion, which achieved a mAP@0.5:0.95 of 0.52 and a mAP@0.5 of 0.94. This result was marginally better than the best baseline (YOLOv5 with a single HMI image) and comparable to other state-of-the-art models, demonstrating a modest but consistent improvement of multispectral image fusion for this task.<\/p>","DOI":"10.21203\/rs.3.rs-7600800\/v1","type":"posted-content","created":{"date-parts":[[2025,9,19]],"date-time":"2025-09-19T06:05:49Z","timestamp":1758261949000},"source":"Crossref","is-referenced-by-count":0,"title":["Multispectral YOLO: Generic Feature Fusion Framework for Solar Active Region Detection"],"prefix":"10.21203","author":[{"given":"Ant\u00f3nio","family":"Santos","sequence":"first","affiliation":[{"name":"LIACC, Faculdade de Engenharia, Universidade do Porto"}]},{"given":"Filipa S.","family":"Barros","sequence":"additional","affiliation":[{"name":"LIACC, Faculdade de Engenharia, Universidade do Porto"}]},{"given":"J. J. G.","family":"Lima","sequence":"additional","affiliation":[{"name":"Instituto de Astrof\u00edsica e Ci\u00eancias do Espa\u00e7o, CAUP"}]},{"given":"Rui F.","family":"Pinto","sequence":"additional","affiliation":[{"name":"Institut de Recherche and Astrophysique et Plan\u00e9tologie, OMP\/CNRS, CNES, University of Toulouse"}]},{"given":"Andr\u00e9","family":"Restivo","sequence":"additional","affiliation":[{"name":"LIACC, Faculdade de Engenharia, Universidade do Porto"}]},{"given":"Lu\u00eds F.","family":"Teixeira","sequence":"additional","affiliation":[{"name":"Instituto de Engenharia de Sistemas e Computadores, Tecnologia e Ci\u00eancia (INESC TEC)"}]}],"member":"297","reference":[{"key":"ref1","doi-asserted-by":"crossref","unstructured":"Schwenn, Rainer (2006) Space weather: The solar perspective. Living reviews in solar physics 3(1): 1--72 Springer","DOI":"10.12942\/lrsp-2006-2"},{"key":"ref2","doi-asserted-by":"crossref","unstructured":"Weber, Maria A and Schunker, Hannah and Jouve, Laur{\\`e}ne and I{\\c{s}}{\\i}k, Emre (2023) Understanding active region origins and emergence on the sun and other cool stars. Space Science Reviews 219(8): 63 Springer","DOI":"10.1007\/s11214-023-01006-5"},{"key":"ref3","doi-asserted-by":"crossref","unstructured":"Abd, Mehmmood A and Majed, Sarab F and Zharkova, V (2010) Automated classification of sunspot groups with support vector machines. Springer, 321--325, Technological developments in networking, education and automation","DOI":"10.1007\/978-90-481-9151-2_56"},{"key":"ref4","doi-asserted-by":"crossref","unstructured":"Almahasneh, Majedaldein and Paiement, Adeline and Xie, Xianghua and Aboudarham, Jean (2022) MSMT-CNN for solar active region detection with multi-spectral analysis. SN Computer Science 3(3): 197 Springer","DOI":"10.1007\/s42979-022-01088-y"},{"key":"ref5","doi-asserted-by":"crossref","unstructured":"Carvalho, Sara and Gomes, S and Barata, Teresa and Louren{\\c{c}}o, A and Peixinho, Nuno (2020) Comparison of automatic methods to detect sunspots in the Coimbra Observatory spectroheliograms. Astronomy and Computing 32: 100385 Elsevier","DOI":"10.1016\/j.ascom.2020.100385"},{"key":"ref6","unstructured":"Gupta, Abhishek (2019) Current research opportunities for image processing and computer vision. Computer Science 20: 387--410 Akademia G{\\'o}rniczo-Hutnicza im. Stanis{\\l}awa Staszica w Krakowie. Wydawnictwo AGH"},{"key":"ref7","doi-asserted-by":"crossref","unstructured":"Santos, Jos{\\'e} and Peixinho, Nuno and Barata, Teresa and Pereira, Carlos and Coimbra, A Paulo and Cris{\\'o}stomo, Manuel M and Mendes, Mateus (2023) Sunspot detection using YOLOv5 in spectroheliograph H-alpha images. Applied Sciences 13(10): 5833 MDPI","DOI":"10.3390\/app13105833"},{"key":"ref8","doi-asserted-by":"crossref","unstructured":"Gallagher, James E and Oughton, Edward J (2025) Surveying You Only Look Once (YOLO) Multispectral Object Detection Advancements, Applications And Challenges. IEEE Access IEEE","DOI":"10.1109\/ACCESS.2025.3526458"},{"key":"ref9","doi-asserted-by":"crossref","unstructured":"Liewer, Paulett C and Neugebauer, M and Zurbuchen, Thomas (2004) Characteristics of active-region sources of solar wind near solar maximum. Solar Physics 223: 209--229 Springer","DOI":"10.1007\/s11207-004-1105-z"},{"key":"ref10","doi-asserted-by":"crossref","unstructured":"Talib, Manar Abu and Majzoub, Sohaib and Nasir, Qassim and Jamal, Dina (2021) A systematic literature review on hardware implementation of artificial intelligence algorithms. The Journal of Supercomputing 77(2): 1897--1938 Springer","DOI":"10.1007\/s11227-020-03325-8"},{"key":"ref11","doi-asserted-by":"crossref","unstructured":"Gong, Long and Yang, Yunfei and Feng, Song and Dai, Wei and Liang, Bo and Xiong, Jianping (2024) Solar active regions detection and tracking based on deep learning. Solar Physics 299(8): 121 Springer","DOI":"10.1007\/s11207-024-02362-3"},{"key":"ref12","unstructured":"Ultralytics Inc. {Ultralytics}. 2025-06-30, https:\/\/www.ultralytics.com\/"},{"key":"ref13","unstructured":"Jocher, Glenn and Stoken, Alex and Borovec, Jirka and Changyu, Liu and Hogan, Adam and Diaconu, Laurentiu and Ingham, Francisco and Poznanski, Jake and Fang, Jiacong and Yu, Lijun and others (2020) ultralytics\/yolov5: v3. 1-bug fixes and performance improvements. Zenodo"},{"key":"ref14","doi-asserted-by":"crossref","unstructured":"Baek, Ji-Hye and Kim, Sujin and Choi, Seonghwan and Park, Jongyeob and Kim, Jihun and Jo, Wonkeun and Kim, Dongil (2021) Solar event detection using deep-learning-based object detection methods. Solar Physics 296(11): 160 Springer","DOI":"10.1007\/s11207-021-01902-5"},{"key":"ref15","unstructured":"Dem{\\v{s}}ar, Janez (2006) Statistical comparisons of classifiers over multiple data sets. Journal of Machine learning research 7(Jan): 1--30"},{"key":"ref16","doi-asserted-by":"crossref","unstructured":"Glogowski, Kolja and Bobra, Monica G and Choudhary, Nitin and Amezcua, Arthur B and Mumford, Stuart J (2019) drms: A python package for accessing hmi and aia data. Journal of Open Source Software 4(40): 1614","DOI":"10.21105\/joss.01614"},{"key":"ref17","doi-asserted-by":"crossref","unstructured":"Barnes, Will T and Bobra, Monica G and Christe, Steven D and Freij, Nabil and Hayes, Laura A and Ireland, Jack and Mumford, Stuart and Perez-Suarez, David and Ryan, Daniel F and Shih, Albert Y and others (2020) The SunPy project: Open source development and status of the version 1.0 core package. The Astrophysical Journal 890(1): 68 IOP Publishing","DOI":"10.3847\/1538-4357\/ab4f7a"},{"key":"ref18","doi-asserted-by":"crossref","unstructured":"Everingham, Mark and Van Gool, Luc and Williams, Christopher KI and Winn, John and Zisserman, Andrew (2010) The pascal visual object classes (voc) challenge. International journal of computer vision 88: 303--338 Springer","DOI":"10.1007\/s11263-009-0275-4"},{"key":"ref19","doi-asserted-by":"crossref","unstructured":"Palladino, Luigi and Ntagiou, Evridiki and Klug, Johannes and Palacios, Judit and Keil, Ralf (2022) Sunspot groups detection and classification on SDO\/HMI images using deep learning techniques. IEEE, 1--10, 2022 IEEE aerospace conference (AERO)","DOI":"10.1109\/AERO53065.2022.9843222"},{"key":"ref20","doi-asserted-by":"crossref","unstructured":"Kucuk, Ahmet and Aydin, Berkay and Angryk, Rafal (2017) Multi-wavelength solar event detection using faster R-CNN. IEEE, 2552--2558, 2017 IEEE International Conference on Big Data (Big Data)","DOI":"10.1109\/BigData.2017.8258214"},{"key":"ref21","doi-asserted-by":"crossref","unstructured":"Banerjee, Chayan and Nguyen, Kien and Fookes, Clinton and George, Karniadakis (2024) Physics-Informed Computer Vision: A Review and Perspectives. ACM Comput. Surv. 57(1): 17:1-17:38 https:\/\/doi.org\/10.1145\/3689037, October, The incorporation of physical information in machine learning frameworks is opening and transforming many application domains. Here the learning process is augmented through the induction of fundamental knowledge and governing physical laws. In this work, we explore their utility for computer vision tasks in interpreting and understanding visual data. We present a systematic literature review of more than 250 papers on formulation and approaches to computer vision tasks guided by physical laws. We begin by decomposing the popular computer vision pipeline into a taxonomy of stages and investigate approaches to incorporate governing physical equations in each stage. Existing approaches are analyzed in terms of modeling and formulation of governing physical processes, including modifying input data (observation bias), network architectures (inductive bias), and training losses (learning bias). The taxonomy offers a unified view of the application of the physics-informed capability, highlighting where physics-informed learning has been conducted and where the gaps and opportunities are. Finally, we highlight open problems and challenges to inform future research. While still in its early days, the study of physics-informed computer vision has the promise to develop better computer vision models that can improve physical plausibility, accuracy, data efficiency, and generalization in increasingly realistic applications., 0360-0300","DOI":"10.1145\/3689037"},{"key":"ref22","doi-asserted-by":"crossref","unstructured":"Wilcoxon, Frank (1945) Individual Comparisons by Ranking Methods. Biometrics Bulletin 1(6): 80 \u201383 https:\/\/doi.org\/10.2307\/3001968, [International Biometric Society, Wiley], 0099-4987","DOI":"10.2307\/3001968"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/www.researchsquare.com\/article\/rs-7600800\/v1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.researchsquare.com\/article\/rs-7600800\/v1.html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T16:15:34Z","timestamp":1776096934000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.researchsquare.com\/article\/rs-7600800\/v1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,19]]},"references-count":22,"URL":"https:\/\/doi.org\/10.21203\/rs.3.rs-7600800\/v1","relation":{"is-preprint-of":[{"id-type":"doi","id":"10.1007\/s00138-026-01807-y","asserted-by":"subject"}]},"subject":[],"published":{"date-parts":[[2025,9,19]]},"subtype":"preprint"}}