GEE数据集——1986—2022年DEA （Digital Earth Australia ）澳大利亚数字地球项目几何中值和绝对偏差中值

本文链接：https://blog.csdn.net/qq_31988139/article/details/137238781

简介

该产品提供统计工具，用于利用澳大利亚数字地球中的 Landsat 7 数据时间序列，提供总体状况的年度图像，以及特定年份某一地区的变化程度。

该产品的大地测量部分提供特定年份的 "平均 "无云图像。地磁图像采用多维中值计算，同时使用卫星图像的所有光谱测量值，以保持测量值之间的关系。

该产品的绝对偏差中值部分使用了三种方差测量方法，每种方法都提供了给定年份的 "二阶 "高维统计综合结果。根据亮度和光谱等因素，这三个方差测量值显示了一个区域与 "平均值 "在 "距离 "上的差异程度：

欧氏距离 (EMAD)
余弦（光谱）距离 (SMAD)
布雷-柯蒂斯相似度 (BCMAD)
它们共同提供了特定年份的景观差异信息，对变化检测应用非常有用。

如需了解更多信息，请参阅 DEA 几何中值和中值绝对偏差大地遥感卫星。

https://cmi.ga.gov.au/data-products/dea/645/dea-geometric-median-and-median-absolute-deviation-landsat

该产品是澳大利亚数字地球计划的一部分

Digital Earth Australia Home | Digital Earth Australia | Geoscience Australia

数据介绍

Catalog Owner

Geoscience Australia

Dataset Availability

2000-01-01T00:00:00Z–2021-01-01T00:00:00Z

Dataset Provider

Geoscience Australia NGIS

Contact

https://www.ga.gov.au/contact-us

Earth Engine Snippet

ee.ImageCollection("projects/geoscience-aus-cat/assets/ga_ls7e_nbart_gm_cyear_3")

Resolution
25 meters

波段

Name	Max	Wavelength	Description
`blue`	10000*	0.450-0.520 μm	Band blue surface reflectance geometric median.
`green`	10000*	0.520-0.600 μm	Band green surface reflectance geometric median.
`red`	10000*	0.630-0.690 μm	Band red surface reflectance geometric median.
`near_infrared`	10000*	0.760-0.900 μm	Band near infrared surface reflectance geometric median.
`shortwave_infrared_1`	10000*	1.550-1.750 μm	Band shortwave infrared 1 surface reflectance geometric median.
`shortwave_infrared_2`	10000*	2.080-2.350 μm	Band shortwave infrared 2 surface reflectance geometric median.
`Euclidean_distance_median_absolute_deviation`	10000*		The Median Absolute Deviation using Euclidean distance (EMAD). EMAD is more sensitive to changes in target brightness.
`spectral_distance_median_absolute_deviation`	10000*		The Median Absolute Deviation using Cosine (spectral) distance (SMAD). SMAD is more sensitive to change in target spectral response.
`Bray_Curtis_dissimilarity_median_absolute_deviation`	10000*		The Median Absolute Deviation using Bray Curtis dissimilarity (BCMAD). BCMAD is more sensitive to the distribution of the observation values through time.
`count`	400*		The number of the available pixels used for calculation per calendar year.

* estimated min or max value

代码

var geomedian_ls7 = ee.ImageCollection('projects/geoscience-aus-cat/assets/ga_ls7e_nbart_gm_cyear_3');

var geometry = 
    /* color: #98ff00 */
    /* displayProperties: [
      {
        "type": "rectangle"
      }
    ] */
    ee.Geometry.Polygon(
      [[[121.15880998755823, -15.010654451073695],
        [121.15880998755823, -18.377531570740548],
        [125.81701311255823, -18.377531570740548],
        [125.81701311255823, -15.010654451073695]]], null, false);
          
var composite = geomedian_ls7.filterBounds(geometry)
                             .filterDate('2008-01-01', '2009-01-01')
                             .mosaic();
                             
var visualization = {
  bands: ['red', 'green', 'blue'],
  min: 0,
  max: 3000
};

Map.centerObject(geometry, 10);
Map.addLayer(composite, visualization, '2008 True Color Composite');

引用

Roberts, D., Mueller, N., & Mcintyre, A. (2017). High-dimensional pixel composites from earth observation time series. IEEE Transactions on Geoscience and Remote Sensing, 55(11), 6254-6264. doi:10.1109/TGRS.2017.2723896. Roberts, D., Dunn, B., & Mueller, N. (2018). Open data cube products using high-dimensional statistics of time series. IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, 8647-8650. doi:10.1109/IGARSS.2018.8518312.