In recent years, remote sensing technology has developed by leaps and bounds. Multiple remote sensing platforms such as aerospace, aviation, and near space have continued to increase. The spatial, temporal, and spectral resolution of data has continued to increase. The amount of data has soared. Remote sensing data has become more and more characteristic of big data. . The emergence of remote sensing big data provides unprecedented opportunities for related research, but how to deal with these data also poses great challenges. Traditional workstations and servers have been unable to meet the needs of large-scale, multi-scale massive remote sensing data processing.
Represented by Earth Engine (GEE) and PIE-Engine, online visualization computing and analysis cloud platforms for global-scale earth science data (especially satellite remote sensing data) are more and more widely used. The GEE platform stores and synchronizes satellite images such as MODIS, Landsat, and Sentinel, climate and weather, and geophysics data sets that are currently commonly used in the field of remote sensing, exceeding 80PB. At the same time, it relies on millions of super servers around the world to provide sufficient computing power for these The data is processed.
Google Earth Engine（GEE）
Google Earth Engine (GEE) is a cloud computing platform jointly developed by Google, Carnegie Mellon University, and the United States Geological Survey (USGS) to process satellite remote sensing image data and other earth observation data.
The GEE platform integrates the powerful computing power provided by Google servers or large-scale cloud computing resources. The platform dataset provides a large number of complete image data of earth observation satellites such as Sentinel, MODIS, Landsat, etc., and also provides vegetation, surface temperature and Social and economic data sets, and the database can be updated every day. GEE provides an editing interface (API) in Python and JavaScript, using a web-based code editor for rapid, interactive algorithm development. It has a particularly outstanding advantage, that is, the amount of data is huge, it can be called online, and the data comes from a wide range of sources, so there is no need to search and download from different source websites according to different data. There is no need to occupy the memory of your own computer, online cloud computing.
Compared with traditional remote sensing image processing tools such as ENVI, GEE has incomparable advantages in processing massive remote sensing data. On the one hand, it provides abundant computing resources, and on the other hand, its huge cloud storage saves a large amount of data download for researchers. And the time of preprocessing, the calculation and analysis visualization of remote sensing data represent the most advanced level in this field in the world, and it is a revolution in the field of remote sensing.
Cases and practices in the field of disasters, water bodies and wetlands
· Introduction of GEE platform and typical application cases;
· Introduction of GEE development environment and commonly used data resources;
Introduction to GPT models such as ChatGPT and Wenxinyiyan
· Introduction to JavaScript basics;
· GEE remote sensing cloud important concepts and typical data analysis process;
GEE basic object introduction, vector and raster object visualization, attribute viewing, API query, basic debugging and other platforms to get started.
GEE basic knowledge interacts with AI models such as ChatGPT
· Basic image operations and operations: mathematical operations, relations/conditions/Boolean operations, morphological filtering, texture feature extraction; image masking, cropping and mosaic, etc.;
· Element basic operations and operations: geometric buffer, intersection, union, difference operations, etc.;
Collection object operations: loop iteration (map/iterate), merge Merge, union (Join);
· Data integration Reduce: including image and image set integration, image synthesis, image area statistics and domain statistics, group integration and area neighborhood statistics, image set linear regression analysis, etc.;
· Machine learning algorithms: including supervised (random forest, CART, SVM, decision tree, etc.) and unsupervised (wekaKMeans, wekaLVQ, etc.) classification algorithms, classification accuracy evaluation, etc.;
Data asset management: including local vector and raster data upload, cloud vector and raster data download, statistical result data export, etc.;
· Drawing visualization: including bar charts, histograms, scatter plots, time series and other graphic drawing.
Demos, including tips for assisted Q&A, code generation and correction.
Mini-case lectures on important knowledge points and GPT model interactive demonstration
1) Landsat, Sentinel-2 images automatically remove clouds and shadows in batches
2) Combine Landsat and Sentinel-2 to calculate vegetation index and annual synthesis in batches
3) Statistical analysis of the number of available images and the number of cloud-free observations in the study area
4) Synthesis of annual NDVI vegetation numbers in China and search for the greenest DOY time of the year
5) Moving window smoothing of time series optical image data
6) Stratified random sampling and sample export, sample local evaluation and data upload to the cloud
Analysis on the Variation Trend of Rainfall in China in the Past 40 Years
Flood Disaster Monitoring
Based on images such as Sentinel-1 radar, a typical flood disaster is taken as an example to monitor the affected area. The case content includes multi-source image data processing and the construction of different water body recognition algorithms, such as OSTU global automatic segmentation and local adaptive threshold method, and the use of different methods to determine the disaster area, disaster area statistics and visual output, etc.
Flood Sensitivity and Risk Modeling
Combining spatial data sets such as ESA10m resolution land cover products, terrain (elevation and slope), MERIT global hydrological data, JRC surface water data products, etc., with the help of cloud platform to calculate the distance between different land types and open waters, the height above the nearest drainage system ( HAND) and rainfall frequency (a representative of rainfall intensity and duration) are used as input parameters for simulating flood sensitivity, and then the weighted linear combination WLC method is used to draw a flood sensitivity distribution map. The content involves reclassification and grading of different data products, Euclidean distance calculation, image set map cycle and analysis modeling, etc.
Water Quality Monitoring
Combined with Landsat 8/9 and JRC surface aquatic products in the past ten years, use such as NDSSI normalized differential suspended sediment index, NDTI normalized differential turbidity index, etc. to monitor water quality changes in water catchment areas, and collect statistics on catchment areas Monthly water quality changes. The content involves time series image preprocessing, vegetation index calculation, monthly and yearly image synthesis, image set Reducer operation, null value filtering and mapping, etc.
River profile monitoring
Demonstrate the application of Earth Engine to river hydrology and geomorphology. Specifically demonstrate how to use the cloud platform to distinguish rivers from other water bodies, perform basic shape analysis, extract the centerline and width of rivers, and detect changes in river shape over time. The content involves open source package calling, RivWidthCloud key code interpretation, time series image processing, water body remote sensing recognition and data export.
Groundwater change monitoring
The use of GRACE satellite observations to assess groundwater storage changes in large river basins is described in detail, including the application of remote sensing estimated total storage anomalies, land surface model output GLDAS, and in situ observations to address groundwater storage invariance. The content involves using GRACE to draw changes in total water storage, water storage trends, and solving changes in groundwater storage in river basins. The practice knowledge points include image set filtering, set Join, map loop, trend analysis, visualization, etc.
Mangrove remote sensing mapping
Combining Sentinel-1/2 multi-source remote sensing images and machine learning algorithms to draw mangrove distribution maps. Topics include optical and radar data processing, machine learning algorithm application, inversion accuracy evaluation, variable importance analysis, result visualization, raster and vector conversion, etc., and will demonstrate how to use mangrove habitat feature information (such as terrain, and sea Connecting, etc.) to fine-tune the classification results and realize the drawing of high-precision distribution maps.