云层覆盖严重制约了遥感影像的观测与应用。包含云、云阴影、雪、水体、陆地5类地物的标准云检测作为遥感影像预处理的重要环节已在国内外主流卫星中广泛应用,然而SDGSAT-1作为可持续发展议程首发卫星,尚未提供标准的逐像素标记产品,限制了其数据的利用与共享。针对SDGSAT-1多光谱波段有限、易出现云雪混淆与云影水体混淆的问题,本文提出了一种多光谱与热红外数据融合的标准云检测方法,利用热红外波段中厚云低温特性提升云雪区分效果,融合GSWO与DEM数据辅助复杂背景识别,并结合形态学与导向滤波优化边界。实验结果表明,云检测IoU达74.33%,整体精度达85.09%。研究有效填补了SDGSAT-1标准云检测数据的空白,为后续数据的自动标注与高精度检测算法研发提供了重要支撑。
Cloud cover seriously restricts the observation and application of remote sensing imagery. Standard cloud detection products, including five types of ground objects (clouds, cloud shadows, snow, water bodies, and land), are an important part of optical image preprocessing and have been widely used in mainstream satellites both domestically and internationally. However, SDGSAT-1, as the first satellite of the Sustainable Development Agenda (SDA), has not yet provided standard pixel-level labeled products, which restricts the utilization and sharing of its data. Aiming at the limitation that SDGSAT-1’s multispectral bands are few, making it prone to confusion between cloud-snow and cloud shadow-water, this paper proposes a standardized cloud detection method by fusing multispectral and thermal infrared (TIR) data. The method utilizes the low-temperature characteristics of thick clouds in the TIR band to enhance cloud-snow differentiation, incorporates GSWO and DEM data to assist in identifying complex backgrounds, and optimizes cloud mask boundaries using morphology and guided filtering. Experimental results show that the cloud detection IoU reaches 74.33% and the overall accuracy reaches 85.09%. The study effectively fills the gap in providing standard cloud detection products for SDGSAT-1 and provides important support for the development of automatic labeling and high-precision detection algorithms for subsequent data.
[1] Zou Z X, Shi Z W.Ship detection in spaceborne optical image with SVD networks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(10): 5832-5845. DOI: 10.1109/TGRS.2016.2572736.
[2] Ma A L, Wang J J, Zhong Y F, et al.FactSeg: Foreground activation-driven small object semantic segmentation in large-scale remote sensing imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-16. DOI: 10.1109/TGRS.2021.3097148.
[3] Chen H, Li W Y, Shi Z W.Adversarial instance augmentation for building change detection in remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-16. DOI: 10.1109/TGRS.2021.3066802.
[4] Guo H D, Dou C Y, Chen H Y, et al.SDGSAT-1: the world’s first scientific satellite for sustainable development goals[J]. Science Bulletin, 2023, 68(1): 34-38. DOI: 10.1016/j.scib.2022.12.014.
[5] Irish R R, Barker J L, Goward S N, et al.Characterization of the Landsat-7 ETM+ automated cloud-cover assessment (ACCA) algorithm[J]. Photogrammetric Engineering and Remote Sensing, 2006, 72(10): 1179-1188. DOI: 10.14358/PERS.72.10.1179.
[6] Zhu Z, Woodcock C E.Object-based cloud and cloud shadow detection in Landsat imagery[J]. Remote Sensing of Environment, 2012, 118: 83-94. DOI: 10.1016/j.rse.2011.10.028.
[7] Luo Y, Trishchenko A P, Khlopenkov K V.Developing clear-sky, cloud and cloud shadow mask for producing clear-sky composites at 250 m spatial resolution for the seven MODIS land bands over Canada and North America[J]. Remote Sensing of Environment, 2008, 112(12): 4167-4185. DOI: 10.1016/j.rse.2008.06.010.
[8] Dong Z, Sun L, Liu X R, et al.CDAG-improved algorithm and its application to GF-6 WFV data cloud detection[J]. Acta Optica Sinica, 2020, 40(16): 1628001. DOI: 10.3788/AOS202040.1628001.
[9] Sun L, Mi X T, Wei J, et al.A cloud detection algorithm-generating method for remote sensing data at visible to short-wave infrared wavelengths[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 124: 70-88. DOI: 10.1016/j.isprsjprs.2016.12.005.
[10] Wang M, Zhang Z, Dong Z, et al.Stream-computing of high accuracy on-board real-time cloud detection for high resolution optical satellite imagery[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(2): 50-59. DOI: 10.11947/j.JGGS.2019.0206.
[11] Li Z W, Shen H F, Li H F, et al.Multi-feature combined cloud and cloud shadow detection in GaoFen-1 wide field of view imagery[J]. Remote Sensing of Environment, 2017, 191: 342-358. DOI: 10.1016/j.rse.2017.01.026.
[12] Hu C M, Zhang Z, Tang P.Research on multispectral satellite image cloud and cloud shadow detection algorithm of domestic satellite[J]. Journal of Remote Sensing, 2023, 27(3): 623-634. DOI: 10.11834/jrs.20211209.
[13] Li Z W, Shen H F, Cheng Q, et al.Deep learning based cloud detection for medium and high resolution remote sensing images of different sensors[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 150: 197-212. DOI: 10.1016/j.isprsjprs.2019.02.017.
[14] Ge K Q, Liu J Y, Wang F, et al.A cloud detection method based on spectral and gradient features for SDGSAT-1 multispectral images[J]. Remote Sensing, 2023, 15(1): 24. DOI: 10.3390/rs15010024.
[15] Pekel J F, Cottam A, Gorelick N, et al.High-resolution mapping of global surface water and its long-term changes[J]. Nature, 2016, 540(7633): 418-422. DOI: 10.1038/nature20584.
[16] Zhang Y, Guindon B, Cihlar J.An image transform to characterize and compensate for spatial variations in thin cloud contamination of Landsat images[J]. Remote Sensing of Environment, 2002, 82(2-3): 173-187. DOI: 10.1016/S0034-4257(02)00034-2.
[17] Gómez-Chova L, Zurita-Milla R, Camps-Valls G, et al.Cloud screening and multitemporal unmixing of MERIS FR data[EB/OL]. Envisat Symposium 2007,(2007-04-23)[2025-09-07].https://earth.esa.int/eogateway/events/envisat-symposium-2007.
[18] Yu W H, Luo M, Zhou P, et al.MetaFormer is actually what you need for vision[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 18-24, 2022, New Orleans, LA, USA. IEEE, 2022: 10819-10829. DOI: 10.1109/CVPR52688.2022.01055.
[19] Wang L B, Fang S H, Meng X L, et al.Building extraction with vision transformer[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-11. DOI: 10.1109/TGRS.2022.3186634.
[20] Strudel R, Garcia R, Laptev I, et al.Segmenter: transformer for semantic segmentation[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV). October 10-17, 2021, Montreal, QC, Canada. IEEE, 2021: 7262-7272. DOI: 10.1109/ICCV48922.2021.00717.
[21] Yuan W, Wang J, Xu W B.Shift pooling PSPNet: rethinking PSPNet for building extraction in remote sensing images from entire local feature pooling[J]. Remote Sensing, 2022, 14(19): 4889. DOI: 10.3390/rs14194889.
[22] Guo M H, Lu C Z, Hou Q, et al.SegNeXt: rethinking convolutional attention design for semantic segmentation[C]//Advances in Neural Information Processing Systems (NeurIPS 2022). 2022, 35: 1140-1156. DOI: 10.48550/arXiv.2209.08575.
