Sentinel-2 Cloud Mask Catalogue

Francis, Alistair; Mrziglod, John; Sidiropoulos, Panagiotis; Müller, Jan-Peter

doi:10.5281/zenodo.4172871

Cited by 9 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1) Sentinel-2 Cloud Mask Catalogue: We produced a dataset [56] comprising 513 subscenes taken from the 2018 archive of Sentinel-2 Level-1C imagery. Sampling of the scenes was completely random, so that the data approximated the real-world distribution of Sentinel-2 data.…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

SEnSeI: A Deep Learning Module for Creating Sensor Independent Cloud Masks

Francis

Mrziglod

Sidiropoulos

et al. 2022

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

We introduce a novel neural network architecture-Spectral ENcoder for SEnsor Independence (SEnSeI)-by which several multispectral instruments, each with different combinations of spectral bands, can be used to train a generalised deep learning model. We focus on the problem of cloud masking, using several pre-existing datasets, and a new, freely available dataset for Sentinel-2. Our model is shown to achieve state-of-the-art performance on the satellites it was trained on (Sentinel-2 and Landsat 8), and is able to extrapolate to sensors it has not seen during training such as Landsat 7, Per úSat-1, and Sentinel-3 SLSTR. Model performance is shown to improve when multiple satellites are used in training, approaching or surpassing the performance of specialised, single-sensor models. This work is motivated by the fact that the remote sensing community has access to data taken with a hugely variety of sensors. This has inevitably led to labelling efforts being undertaken separately for different sensors, which limits the performance of deep learning models, given their need for huge training sets to perform optimally. Sensor independence can enable deep learning models to utilise multiple datasets for training simultaneously, boosting performance and making them much more widely applicable. This may lead to deep learning approaches being used more frequently for on-board applications and in ground segment data processing, which generally require models to be ready at launch or soon afterwards.

show abstract

Section: Datasetsmentioning

confidence: 99%

“…The total agreement between the annotators was 95.0%. The dataset is publicly available, and more information can be found in its documentation [56]. In our experiments, we split the dataset in a 40:10:50 ratio, for training, validation and testing respectively.…”

Section: Datasetsmentioning

confidence: 99%

SEnSeI: A Deep Learning Module for Creating Sensor Independent Cloud Masks

Francis

Mrziglod

Sidiropoulos

et al. 2022

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The recent progress in DL-based cloud semantic segmentation in Sentinel-2 can be attributed to the proliferation of public CD datasets such as SPARCS 27 , S2-Hollstein 28 , Biome 8 10 , 38-cloud 29 , CESBIO 30 , 95-Cloud 31 , and CloudCatalogue 32 .…”

Section: Background and Summarymentioning

confidence: 99%

CloudSEN12 - a global dataset for semantic understanding of cloud and cloud shadow in Sentinel-2

Aybar¹,

Ysuhuaylas²,

Loja³

et al. 2022

Preprint

View full text Add to dashboard Cite

Accurately characterizing clouds and their shadows is a long-standing problem in the Earth Observation community. Recent works showcase the necessity to improve cloud detection methods for imagery acquired by the Sentinel-2 satellites. However, the lack of consensus and transparency in existing reference datasets hampers the benchmarking of current cloud detection methods. Exploiting the analysis-ready data offered by the Copernicus program, we created CloudSEN12, a new multi-temporal global dataset to foster research in cloud and cloud shadow detection. CloudSEN12 has 49,400 image patches, including (1) Sentinel-2 level-1C and level-2A multi-spectral data, (2) Sentinel-1 synthetic aperture radar data, (3) auxiliary remote sensing products, (4) different hand-crafted annotations to label the presence of thick and thin clouds and cloud shadows, and (5) the results from eight state-of-the-art cloud detection algorithms. At present, CloudSEN12 exceeds all previous efforts in terms of annotation richness, scene variability, geographic distribution, metadata complexity, quality control, and number of samples. The dataset is made publicly available at https://cloudsen12.github.io/.

show abstract

“…We employed the Cloud Mask Catalogue [22], which contains cloud masks for 513 Sentinel-2A [20] data cubes collected from a variety of geographical regions, each with 13 spectral bands and 20 m ground resolution (1024×1024 pixels). Following Giuffrida et al, who also used Sentinel-2A data, we applied the Level-1C processed version of the data, i.e., top-of-atmosphere reflectance data cubes.…”

Section: Datasetmentioning

confidence: 99%

“…3), we collected Sentinel-2A Level-1C data products [18] over the globe with a distribution of surface types that resembles the Hollstein dataset [27]. The downloaded data cubes were preprocessed following [22], including spatial resampling to achieve a ground resolution of 20 m and size 512×512×13. Sen2Cor [43] was applied to produce probabilistic cloud masks, and a threshold of 0.35 was applied on the probabilities to decide cloudy and not cloudy pixels.…”

Section: Data Collection For Cube Optimisationmentioning

confidence: 99%

Adversarial Attacks against a Satellite-borne Multispectral Cloud Detector

Du¹,

Law²,

Sasdelli³

et al. 2021

Preprint

View full text Add to dashboard Cite

Data collected by Earth-observing (EO) satellites are often afflicted by cloud cover. Detecting the presence of clouds-which is increasingly done using deep learning-is crucial preprocessing in EO applications. In fact, advanced EO satellites perform deep learning-based cloud detection on board the satellites and downlink only clear-sky data to save precious bandwidth. In this paper, we highlight the vulnerability of deep learning-based cloud detection towards adversarial attacks. By optimising an adversarial pattern and superimposing it into a cloudless scene, we bias the neural network into detecting clouds in the scene. Since the input spectra of cloud detectors include the nonvisible bands, we generated our attacks in the multispectral domain. This opens up the potential of multi-objective attacks, specifically, adversarial biasing in the cloud-sensitive bands and visual camouflage in the visible bands. We also investigated mitigation strategies against the adversarial attacks. We hope our work further builds awareness of the potential of adversarial attacks in the EO community.

show abstract

Sentinel-2 Cloud Mask Catalogue

Cited by 9 publications

References 0 publications

SEnSeI: A Deep Learning Module for Creating Sensor Independent Cloud Masks

SEnSeI: A Deep Learning Module for Creating Sensor Independent Cloud Masks

CloudSEN12 - a global dataset for semantic understanding of cloud and cloud shadow in Sentinel-2

Adversarial Attacks against a Satellite-borne Multispectral Cloud Detector

Contact Info

Product

Resources

About