当前位置:网站首页>Google Earth Engine(GEE)——基于s2影像的实时全球10米土地利用/土地覆盖(LULC)数据集
Google Earth Engine(GEE)——基于s2影像的实时全球10米土地利用/土地覆盖(LULC)数据集
2022-06-10 13:02:00 【此星光明2021年博客之星云计算Top3】
全球动态土地分类数据集是一个10米的近实时(NRT)土地利用/土地覆盖(LULC)数据集,包括九个类别的概率和标签信息。
动态世界的预测适用于2015-06-27至今的Sentinel-2 L1C集合。Sentinel-2的重访频率为2-5天,取决于纬度。动态世界预测是针对CLOUDY_PIXEL_PERCENTAGE<=35%的Sentinel-2 L1C图像生成的。预测结果被屏蔽,以去除云和云的阴影,使用S2云的概率、云的位移指数和方向距离变换的组合。
动态世界集合中的图像的名称与它们所来自的单个Sentinel-2 L1C资产名称一致,例如
ee.Image('COPERNICUS/S2/20160711T084022_20160711T084751_T35PKT')
有一个匹配的动态世界图像,名为:ee.Image('GOOGLE/DYNAMICWORLD/V1/20160711T084022_20160711T084751_T35PKT') 。
除了 "标签 "带,所有的概率带总和为1。
要了解更多关于 "动态世界 "数据集的信息,并查看生成合成物、计算区域统计数据和处理时间序列的例子,请参见 "动态世界 "系列教程简介。
鉴于 "动态世界 "的等级估计是通过一个小的移动窗口的空间背景从单一的图像中得出的,因此,如果没有明显的区分特征,预测的土地覆盖物的top-1 "概率 "就会相对较低,这部分是由随时间变化的覆盖物定义的,比如农作物。干旱气候下的高回报率表面、沙子、太阳光等也可能表现出这种现象。
为了只选择有把握的属于动态世界类别的像素,建议通过对前1名预测的估计 "概率 "进行阈值处理来屏蔽动态世界输出。
Dataset Availability
2015-06-23T00:00:00 -
Dataset Provider
World Resources Institute Google
Collection Snippet
ee.ImageCollection("GOOGLE/DYNAMICWORLD/V1")
Resolution
10 meters
Bands Table
| Name | Description | Min | Max |
|---|---|---|---|
| water | Estimated probability of complete coverage by water | 0 | 1 |
| trees | Estimated probability of complete coverage by trees | 0 | 1 |
| grass | Estimated probability of complete coverage by grass | 0 | 1 |
| flooded_vegetation | Estimated probability of complete coverage by flooded vegetation | 0 | 1 |
| crops | Estimated probability of complete coverage by crops | 0 | 1 |
| shrub_and_scrub | Estimated probability of complete coverage by shrub and scrub | 0 | 1 |
| built | Estimated probability of complete coverage by built | 0 | 1 |
| bare | Estimated probability of complete coverage by bare | 0 | 1 |
| snow_and_ice | Estimated probability of complete coverage by snow and ice | 0 | 1 |
| label | Index of the band with the highest estimated probability | 0 | 8 |
Class Table: label
| Value | Color | Color Value | Description |
|---|---|---|---|
| 0 | #419BDF | water | |
| 1 | #397D49 | trees | |
| 2 | #88B053 | grass | |
| 3 | #7A87C6 | flooded_vegetation | |
| 4 | #E49635 | crops | |
| 5 | #DFC35A | shrub_and_scrub | |
| 6 | #C4281B | built | |
| 7 | #A59B8F | bare | |
| 8 | #B39FE1 | snow_and_ice |
属性:
| Name | Type | Description |
|---|---|---|
| dynamicworld_algorithm_version | String | The version string uniquely identifying the Dynamic World model and inference process used to produce the image. |
| qa_algorithm_version | String | The version string uniquely identifying the cloud masking process used to produce the image. |
其它官网链接APP:
App: https://www.dynamicworld.app
论文链接:
https://doi.org/10.1038/s41597-022-01307-4
代码:
// 构建一个相应的动态世界和Sentinel-2的集合,以供检查。按地区和日期过滤DW和S2的集合。
var COL_FILTER = ee.Filter.and(
ee.Filter.bounds(ee.Geometry.Point(20.6729, 52.4305)),
ee.Filter.date('2021-04-02', '2021-04-03'));
//两个数据集按照同样的结果进行筛选
var dwCol = ee.ImageCollection('GOOGLE/DYNAMICWORLD/V1').filter(COL_FILTER);
var s2Col = ee.ImageCollection('COPERNICUS/S2').filter(COL_FILTER);
// 加入相应的DW和S2图像(通过系统:索引)。
var DwS2Col = ee.Join.saveFirst('s2_img').apply(dwCol, s2Col,
ee.Filter.equals({leftField: 'system:index', rightField: 'system:index'}));
// 提取一个DW图像的例子和它的来源S2图像。
var dwImage = ee.Image(DwS2Col.first());
var s2Image = ee.Image(dwImage.get('s2_img'));
//创建一个将DW类标签与概率相融合的可视化。定义DW LULC标签和颜色的列表对。
var CLASS_NAMES = [
'water', 'trees', 'grass', 'flooded_vegetation', 'crops',
'shrub_and_scrub', 'built', 'bare', 'snow_and_ice'];
var VIS_PALETTE = [
'419BDF', '397D49', '88B053', '7A87C6',
'E49635', 'DFC35A', 'C4281B', 'A59B8F',
'B39FE1'];
// 在[0, 1]上创建一个标签(最可能的类别)的RGB图像。
var dwRgb = dwImage
.select('label')
.visualize({min: 0, max: 8, palette: VIS_PALETTE})
.divide(255);
// 获得最可能的分类概率。
var top1Prob = dwImage.select(CLASS_NAMES).reduce(ee.Reducer.max());
// 在[0, 1]上创建一个最可能的类别概率的山形图。
var top1ProbHillshade =
ee.Terrain.hillshade(top1Prob.multiply(100))
.divide(255);
// 将RGB图像与丘陵地带结合起来。
var dwRgbHillshade = dwRgb.multiply(top1ProbHillshade);
// 用源Sentinel-2图像显示动态世界的可视化。
Map.setCenter(20.6729, 52.4305, 12);
Map.addLayer(
s2Image,
{min: 0, max: 3000, bands: ['B4', 'B3', 'B2']},
'Sentinel-2 L1C');
Map.addLayer(
dwRgbHillshade,
{min: 0, max: 0.65},
'Dynamic World');
RGB影像
土地分类结果
北京区域:

边栏推荐
- Buuctf [glassfish] arbitrary file reading
- 【深度学习】基于深度学习Autoencoder的信用卡欺诈异常检测,效果非常牛逼
- Notes - simple but adequate series_ The Yapi return parameter data should be an object type problem solving record
- Introduction to assembly language - Summary
- Which EDA design software should Altium Allegro pads choose
- 不吐不快
- Shi Yigong and other teams posted on the cover of Science: AI and freeze electron microscope revealed the structure of "atomic level" NPC, a breakthrough in life science
- 常见的自动化测试框架有哪些?上海软件测试公司安利
- 十款好用跨浏览器测试工具分享,好物值得收藏
- Six stone programming: talking about naming based on the position of word processing
猜你喜欢

Apple邮箱配置QQ邮箱,163邮箱,edu邮箱,gmail邮箱,获取gmail日历
![buuctf [PHP]XDebug RCE](/img/e2/bcae10e2051b7e9dce918bf87fdc05.png)
buuctf [PHP]XDebug RCE
![buuctf [PHP]inclusion](/img/02/d328ed84e4641c09c5b1eba3ac6ea9.png)
buuctf [PHP]inclusion

Leetcode 96. Différents arbres de recherche binaires

Z-Wave ecosystem status report in 2022
![[NLP] NLP full path learning recommendation](/img/d8/a367c26b51d9dbaf53bf4fe2a13917.png)
[NLP] NLP full path learning recommendation

client-go gin的简单整合六-list-watch二(关于Rs与Pod以及Deployment的完善)
![[Netease Yunxin] in depth analysis of the design of](/img/17/7c0b7db01310f8dd31e9021f8d5516.png)
[Netease Yunxin] in depth analysis of the design of "circle group" relationship system | series of articles on "circle group" technology

拷贝和删除文件

Simple integration of client go gin six list watch two (about the improvement of RS, pod and deployment)
随机推荐
CF894C Marco and GCD Sequence
'getWidth()' is deprecated,'getHeight()' is deprecated
Error:top-left corner pixel must be either opaque white or transparent.
How can the team be dissolved...
In depth analysis of "circle group" relationship system design | series of articles on "circle group" technology
Periodically brush the data in the database and synchronize only the modified fields
Neuron newsletter 2022-05 | add 2 southbound drivers and 1 northbound application, and realize customized extension with Modbus TCP
Ekuiper newsletter 2022-05 protobuf codec support, visual drag and drop writing rules
【Golang】创建有配置参数的结构体时,可选参数应该怎么传?
Z-Wave ecosystem status report in 2022
RecyclerView多布局写法,“我的”、“个人中心” 页面经典写法演示
Development trend of Web Development
The deep neural network classifies nearly 2billion images per second, and the new brain like optical classifier chip is on nature
buuctf [Jupyter]notebook-rce
[NLP] NLP full path learning recommendation
出海企业遇瓶颈 茄子科技(SHAREit Group)有话说
Google proposed the super pre training model coca, and the accuracy of fine-tuning top-1 on Imagenet reached 91%! SOTA on multiple downstream tasks!
Yyds dry goods inventory # solve the problem of sword finger offer: jump step expansion
How does the API detect security configuration errors?
The essence of linear algebra 6 inverse matrix, column space and zero space