Large-scale demonstration of machine learning for the detection of volcanic deformation in Sentinel-1 satellite imagery.
Large-scale demonstration of machine learning for the detection of volcanic deformation in Sentinel-1 satellite imagery.
复制标题
DOI：
10.1007/s00445-022-01608-x
复制
发表时间：
2022
影响因子：
3.5
通讯作者：
中科院分区：
文献类型：
--
作者：
雷达(SAR)卫星系统地获取可用于火山监测、描述岩浆系统特征和潜在地预测全球范围内喷发的图像。然而，利用大数据集受到人工检查需求的限制，这意味着信息的及时传播是具有挑战性的。在这里，我们自动处理由哨兵一号卫星在5年内(2015年至2020年)获得的 ~ 600,000张 >； 1000火山的图像，并使用该数据集演示机器学习在检测变形信号方面的适用性和局限性。在最常见的16座火山中，有5座火山经历了喷发，6座火山变形缓慢，2座火山非火山变形，3座大气人工制品。整个数据集的检测阈值为5.9厘米，相当于5年研究期内1.2厘米/年的速度。然后，我们使用大型测试数据集来探索大气条件、土地覆盖和信号特征对可检测性的影响，发现机器学习算法的性能主要受可用数据质量的限制，一致性差，信号慢尤其具有挑战性。系统获取、处理和标记图像的数据集不断扩大，将能够以前所未有的规模对火山监测信号进行量化分析，但常规监测应用将需要量身定做的处理。网上版载有补充材料，可在10.1007/s00445-022-01608-x查阅。
Radar (SAR) satellites systematically acquire imagery that can be used for volcano monitoring, characterising magmatic systems and potentially forecasting eruptions on a global scale. However, exploiting the large dataset is limited by the need for manual inspection, meaning timely dissemination of information is challenging. Here we automatically process ~ 600,000 images of > 1000 volcanoes acquired by the Sentinel-1 satellite in a 5-year period (2015–2020) and use the dataset to demonstrate the applicability and limitations of machine learning for detecting deformation signals. Of the 16 volcanoes flagged most often, 5 experienced eruptions, 6 showed slow deformation, 2 had non-volcanic deformation and 3 had atmospheric artefacts. The detection threshold for the whole dataset is 5.9 cm, equivalent to a rate of 1.2 cm/year over the 5-year study period. We then use the large testing dataset to explore the effects of atmospheric conditions, land cover and signal characteristics on detectability and find that the performance of the machine learning algorithm is primarily limited by the quality of the available data, with poor coherence and slow signals being particularly challenging. The expanding dataset of systematically acquired, processed and flagged images will enable the quantitative analysis of volcanic monitoring signals on an unprecedented scale, but tailored processing will be needed for routine monitoring applications. The online version contains supplementary material available at 10.1007/s00445-022-01608-x.
登录
查看更多内容
影响因子：
16.6
作者：
Biggs, J.;Ebmeier, S. K.;Aspinall, W. P.;Lu, Z.;Pritchard, M. E.;Sparks, R. S. J.;Mather, T. A.
通讯作者：
Mather, T. A.
影响因子：
18.3
作者：
Flovenz, Olafur G.;Wang, Rongjiang;Milkereit, Claus
通讯作者：
Milkereit, Claus
DOI：
10.1144/sp380.2
发表时间：
2013-01-01
期刊：
REMOTE SENSING OF VOLCANOES AND VOLCANIC PROCESSES: INTEGRATING OBSERVATION AND MODELLING
影响因子：
--
作者：
Ebmeier, S. K.;Biggs, J.;Amelung, F.
通讯作者：
Amelung, F.
影响因子：
3.9
作者：
Beauducel, F;Briole, P;Froger, JL
通讯作者：
Froger, JL
影响因子：
16.6
作者：
Albino, Fabien;Biggs, Juliet;Syahbana, Devy Kamil
通讯作者：
Syahbana, Devy Kamil