World Settlement Footprint (WSF) 2015는 2014~2015년 다중 시간 Landsat-8 및 Sentinel-1 이미지 (각각 약 217,000개 및 약 107,000개 장면 처리)를 사용하여 전 세계 인적 거주지의 범위를 나타내는 10m 해상도 바이너리 마스크입니다.
시간에 따른 인적 거주지의 시간적 역학은 다른 모든 비거주지 정보 클래스와는 다릅니다.
따라서 선택한 시간 간격에서 관심 지역에 사용할 수 있는 모든 다중 시간 이미지에 대해 다음 항목의 주요 시간 통계 (즉, 시간 평균, 최소, 최대 등)가 추출됩니다.
레이더 데이터의 경우 원래 역산란 값
광학 이미지의 경우 클라우드 마스크를 수행한 후 파생된 다양한 스펙트럼 지수 (예: 식생 지수, 건물 지수 등)
그런 다음 지원 벡터 머신 (SVM)을 기반으로 하는 다양한 분류 체계가 광학 및 레이더 시간적 특징에 각각 별도로 적용되고, 마지막으로 두 출력이 적절하게 결합됩니다.
레이어의 높은 정확성과 신뢰성을 정량적으로 평가하기 위해 Google과 협력하여 방대한 양의 그라운드 트루스 샘플 (예: 900,000개)이 크라우드소싱 사진 해석을 통해 라벨이 지정되었습니다. 통계적으로 강력하고 투명한 프로토콜은 현재 문헌에서 권장하는 최신 사례에 따라 정의되었습니다.
Marconcini, M., Metz-Marconcini, A., Üreyen, S., Palacios-Lopez, D., Hanke, W., Bachofer, F.,
Zeidler, J., Esch, T., Gorelick, N., Kakarla, A., Paganini, M., Strano, E. (2020).
인간이 거주하는 지역을 나타내는 World Settlement Footprint 2015. Scientific Data, 7(1), 1~14.
doi:10.1038/s41597-020-00580-5
World Settlement Footprint (WSF) 2015는 2014~2015년 다중 시간 Landsat-8 및 Sentinel-1 이미지 (각각 약 217,000개 및 약 107,000개의 장면이 처리됨)를 통해 전 세계적으로 인간 정착지의 범위를 나타내는 10m 해상도 이진 마스크입니다. 시간이 지남에 따른 인간 정착지의 시간적 역학은 …
[null,null,[],[[["\u003cp\u003eThe World Settlement Footprint (WSF) 2015 dataset provides a 10m resolution global map of human settlements, derived from 2014-2015 Landsat-8 and Sentinel-1 imagery.\u003c/p\u003e\n"],["\u003cp\u003eIt utilizes temporal dynamics and spectral indices to distinguish settlement areas from other land cover types with high accuracy, validated by extensive ground-truth samples.\u003c/p\u003e\n"],["\u003cp\u003eThe dataset is freely available under a CC0-1.0 license and can be accessed and analyzed through Google Earth Engine.\u003c/p\u003e\n"],["\u003cp\u003eCreated by the German Aerospace Center (DLR) in collaboration with Google, the WSF 2015 is detailed in a Scientific Data publication.\u003c/p\u003e\n"]]],["The World Settlement Footprint (WSF) 2015 dataset, provided by DLR, is a global 10m resolution binary mask of human settlements. It uses 2014-2015 Landsat-8 and Sentinel-1 imagery, processing approximately 217,000 and 107,000 scenes, respectively. Temporal statistics from radar and optical data, including spectral indices, are extracted. Support Vector Machines (SVMs) are applied to classify settlements, with outputs combined and validated using 900,000 ground-truth samples. The dataset is accessible via Earth Engine.\n"],null,["# World Settlement Footprint 2015\n\nDataset Availability\n: 2015-01-01T00:00:00Z--2016-01-01T00:00:00Z\n\nDataset Provider\n:\n\n\n [Deutsches Zentrum für Luft- und Raumfahrt (DLR)](https://www.dlr.de/)\n\nTags\n:\n[landcover](/earth-engine/datasets/tags/landcover) [landsat-derived](/earth-engine/datasets/tags/landsat-derived) [population](/earth-engine/datasets/tags/population) [sentinel1-derived](/earth-engine/datasets/tags/sentinel1-derived) [settlement](/earth-engine/datasets/tags/settlement) [urban](/earth-engine/datasets/tags/urban) \n\n#### Description\n\nThe World Settlement Footprint (WSF) 2015 is a 10m resolution binary mask\noutlining the extent of human settlements globally derived by means of\n2014-2015 multitemporal Landsat-8 and Sentinel-1 imagery (of which \\~217,000 and\n\\~107,000 scenes have been processed, respectively).\n\nThe temporal dynamics of human settlements over time are\nsensibly different than those of all other non-settlement information classes.\nHence, given all the multitemporal images available over a region of interest\nin the selected time interval, key temporal statistics (i.e., temporal mean,\nminimum, maximum, etc.) are extracted for:\n\n- the original backscattering value in the case of radar data; and\n- different spectral indices (e.g., vegetation index, built-up index, etc.) derived after performing cloud masking in the case of optical imagery.\n\nNext, different classification schemes based on Support\nVector Machines (SVMs) are separately applied to the optical and radar temporal\nfeatures, respectively, and, finally, the two outputs are properly combined\ntogether.\n\nTo quantitatively assess the high accuracy and reliability of the\nlayer, an extensive validation exercise has been carried out in collaboration\nwith Google based on a huge amount of ground-truth samples (i.e., 900,000)\nlabeled by crowd-sourcing photo-interpretation. A statistically\nrobust and transparent protocol has been defined following the state-of-the-art\npractices currently recommended in the literature.\n\nFor all technical details, please refer to\n[the publication](https://www.nature.com/articles/s41597-020-00580-5)\n\n### Bands\n\n**Bands**\n\n| Name | Min | Max | Pixel Size | Description |\n|--------------|-----|-----|------------|-------------------------|\n| `settlement` | 255 | 255 | 10 meters | A human settlement area |\n\n### Terms of Use\n\n**Terms of Use**\n\n[CC0-1.0](https://spdx.org/licenses/CC0-1.0.html)\n\n### Citations\n\nCitations:\n\n- Marconcini, M., Metz-Marconcini, A., Üreyen, S., Palacios-Lopez, D., Hanke, W., Bachofer, F.,\n Zeidler, J., Esch, T., Gorelick, N., Kakarla, A., Paganini, M., Strano, E. (2020).\n Outlining where humans live, the World Settlement Footprint 2015. Scientific Data, 7(1), 1-14.\n [doi:10.1038/s41597-020-00580-5](https://doi.org/10.1038/s41597-020-00580-5)\n\n### DOIs\n\n- \u003chttps://doi.org/10.1038/s41597-020-00580-5\u003e\n\n### Explore with Earth Engine\n\n| **Important:** Earth Engine is a platform for petabyte-scale scientific analysis and visualization of geospatial datasets, both for public benefit and for business and government users. Earth Engine is free to use for research, education, and nonprofit use. To get started, please [register for Earth Engine access.](https://console.cloud.google.com/earth-engine)\n\n### Code Editor (JavaScript)\n\n```javascript\nvar dataset = ee.Image('DLR/WSF/WSF2015/v1');\n\nvar opacity = 0.75;\nvar blackBackground = ee.Image(0);\nMap.addLayer(blackBackground, null, 'Black background', true, opacity);\n\nvar visualization = {\n min: 0,\n max: 255,\n};\nMap.addLayer(dataset, visualization, 'Human settlement areas');\n\nMap.setCenter(90.45, 23.7, 7);\n```\n[Open in Code Editor](https://code.earthengine.google.com/?scriptPath=Examples:Datasets/DLR/DLR_WSF_WSF2015_v1) \n[World Settlement Footprint 2015](/earth-engine/datasets/catalog/DLR_WSF_WSF2015_v1) \nThe World Settlement Footprint (WSF) 2015 is a 10m resolution binary mask outlining the extent of human settlements globally derived by means of 2014-2015 multitemporal Landsat-8 and Sentinel-1 imagery (of which \\~217,000 and \\~107,000 scenes have been processed, respectively). The temporal dynamics of human settlements over time are sensibly different ... \nDLR/WSF/WSF2015/v1, landcover,landsat-derived,population,sentinel1-derived,settlement,urban \n2015-01-01T00:00:00Z/2016-01-01T00:00:00Z \n-90 -180 90 180 \nGoogle Earth Engine \nhttps://developers.google.com/earth-engine/datasets\n\n- [https://doi.org/10.1038/s41597-020-00580-5](https://doi.org/https://www.dlr.de/)\n- [https://doi.org/10.1038/s41597-020-00580-5](https://doi.org/https://developers.google.com/earth-engine/datasets/catalog/DLR_WSF_WSF2015_v1)"]]
