|Project Name||Stars||Downloads||Repos Using This||Packages Using This||Most Recent Commit||Total Releases||Latest Release||Open Issues||License||Language|
|Gdal||4,183||60||11 hours ago||36||August 04, 2023||435||other||C++|
|GDAL is an open source MIT licensed translator library for raster and vector geospatial data formats.|
|Datacube Core||461||7||18||3 days ago||41||November 08, 2023||41||apache-2.0||Python|
|Open Data Cube analyses continental scale Earth Observation data through time|
|Pyrosar||443||2 days ago||32||November 23, 2023||44||mit||Python|
|framework for large-scale SAR satellite data processing|
|Waterdetect||160||1||6 months ago||22||October 02, 2022||1||apache-2.0||Jupyter Notebook|
|Water Detect Algorithm|
|Open Geo Tutorial||158||4 years ago||4||other||HTML|
|Tutorial of basic remote sensing and GIS methodologies using open source software (GDAL in Python or R)|
|Geowombat||152||7 days ago||15||October 11, 2022||8||mit||Jupyter Notebook|
|GeoWombat: Utilities for geospatial data|
|Modistsp||146||2 months ago||19||June 12, 2023||3||gpl-3.0||R|
|An "R" package for automatic download and preprocessing of MODIS Land Products Time Series|
|Tomography_tutorial||81||8 months ago||4||September 11, 2018||1||gpl-3.0||Jupyter Notebook|
|A tutorial for Synthetic Aperture Radar Tomography|
|Digital Elevation Model||52||3 years ago||1||mit||Python|
|To transform, project, visualize, and read Digital Elevation Models (ASTERGDEMv2.0 and EUDEMv1.1). [简单实用] 数字高程模型的变换、投影、可视化和读取。|
|Label Pixels||42||a year ago||1||Python|
|Label-Pixels is the tool for semantic segmentation of remote sensing images using Fully Convolutional Networks. Initially, it is designed for extracting the road network from remote sensing imagery and now, it can be used to extract different features from remote sensing imagery.|
WaterDetect is an end-to-end algorithm to generate open water cover mask, specially conceived for L2A Sentinel 2 imagery from MAJA1 processor, without any a priori knowledge on the scene. It can also be used for Landsat 8 images and for other multispectral clustering/segmentation tasks.
The water masks produced by WaterDetect were primarily designed for water quality product computation (Obs2Co processing chain) and are also used for multi-temporal water maps (Surfwater processing chain). Both chains are supported by the "SWOT-Downstream" and TOSCA programs by CNES. Products are provided by the THEIA / Hydroweb-NG platform.
The WaterDetect algorithm uses a multidimensional agglomerative clustering technique on a subsample of the scene's pixels, to group them in classes, and a naive bayes classifier to generalize the results for the whole scene, as summarized in the following picture:
All the details and tests has been described in the article Automatic Water Detection from Multidimensional Hierarchical Clustering for Sentinel-2 Images and a Comparison with Level 2A Processors, under revision by the journal Remote Sensing of Environment.
Cordeiro, M. C. R.; Martinez, J.-M.; Peña-Luque, S. Automatic Water Detection from Multidimensional Hierarchical Clustering for Sentinel-2 Images and a Comparison with Level 2A Processors. Remote Sensing of Environment 2021, 253, 112209. https://doi.org/10.1016/j.rse.2020.112209.
waterqualitypackage. For more information, check the waterquality package here: cordmaur/WaterQuality
process_ext_masks. It prepares an external mask (ex. created by FMask) to be used by WaterDetect.
The following topics have the first steps to install and run the library. For a more comprehensive tutorial with code samples and results please refer to this tutorial https://cordmaur.medium.com/water-detection-in-high-resolution-satellite-images-using-the-waterdetect-python-package-7c5a031e3d16.
The algorithm has been developed taking into account atmospherically corrected images from MAJA, as described in the paper. However other image formats are also supported. To the present, the following image formats are supported:
The required libraries are:
GDAL>=3.0.2 matplotlib>=3.1.2 PyPDF2>=1.26.0 scipy>=1.3.2 scikit-learn>=0.22 scikit-image>=0.16.2 numpy>=1.17 PIL>=8.0 lxml>=4.5
GDAL is required to open the satellite images. It's still possible to use without GDAL, from a python console or jupyter notebook, loading the rasters manually and passing all the necessary bands to the DWImageClustering class. Check the topic "Usage from Console" for more information.
Scikit-Image is only necessary to run Otsu threshold method.
The test_dependencies.py can be used to check if all libraries are loading correctly. Simply run:
The easiest way to install waterdetect package is with
pip install waterdetect
Alternatively, you can clone the repository and install from its root throught the following commands:
git clone https://github.com/cordmaur/WaterDetect.git cd WaterDetect pip install .
Once installed, a
waterdetect entry point is created in the path of the environment.
One can check the installation and options by running
waterdetect --help. If GDAL is not found, a message will raise indicating that waterdetect will only run from a console.
usage: waterdetect [-h] [-GC] [-i INPUT] [-o OUT] [-s SHP] [-p PRODUCT] [-c CONFIG] The waterdetect is a high speed water detection algorithm for satellite images. It will loop through all images available in the input folder and write results for every combination specified in the .ini file to the output folder. It can also run for single images from Python console or Jupyter notebook. Refer to the onlinedocumentation optional arguments: -h, --help show this help message and exit -GC, --GetConfig Copy the WaterDetect.ini from the package into the specifieddirectory and skips the processing. Once copied you can edit the .ini file and launch the waterdetect without -c option. -i INPUT, --input INPUT The products input folder. Required. -o OUT, --out OUT Output directory. Required. -s SHP, --shp SHP SHP file. Optional. -p PRODUCT, --product PRODUCT The product to be processed (S2_THEIA, L8_USGS, S2_L1C or S2_S2COR) -c CONFIG, --config CONFIG Configuration .ini file. If not specified WaterDetect.ini from current dir and used as default To copy the package's default .ini file into the current directory, type: `waterdetect -GC .` without other arguments and it will copy WaterDetect.ini into the current directory.
The waterdetect needs a config file that specifies the bands used in the clustering process as well as other parameters.
To obtain the default version of this file, one can use
waterdetec -GC and the file WaterDetect.ini will be copied into the current working folder.
The basic usage for the waterdetect is:
waterdetect -i c:/input_folder -o -c:/output_folder -p S2_THEIA
The input directory should contain the uncompressed folders for the images. The script will loop through all folders in the input directory and save the water masks, graphs and reports to the output folder. The output folder must be created beforehand.
If the config file is not specified, the script will search for WaterDetect.ini in the current folder.
Once properly installed, the WaterDetect can be run from a console or a Jupyter Notebook, by importing the package and calling DWDetectWater.
>>> import waterdetect as wd >>> !waterdetect -GC >>> wd.DWWaterDetect.run_water_detect(input_folder='D:\Images\Input\', output_folder='D:\Images\Output', shape_file='D:\Shp\SomeShapefile.shp', single_mode=False, product=wd.DWProducts.Sentinel2_THEIA, config_file='WaterDetect.ini' )
For more information on how to use it from jupyter notebook, in batch or single mode or to use it with other satellite images or without GDAL, please refer to the tutorial available here https://towardsdatascience.com/water-detection-in-high-resolution-satellite-images-using-the-waterdetect-python-package-7c5a031e3d16.
Author: Maurício Cordeiro (ANA/GET)
Supervisor: Jean-Michel Martinez (IRD/GET)
Validation dataset: Santiago Pena Luque (CNES)
(1) Hagolle, O.; Huc, M.; Pascual, D. V.; Dedieu, G. A Multi-Temporal Method for Cloud Detection, Applied to FORMOSAT-2, VENµS, LANDSAT and SENTINEL-2 Images. Remote Sensing of Environment 2010, 114 (8), 1747–1755. https://doi.org/10.1016/j.rse.2010.03.002.
(2) Cordeiro, M. C. R.; Martinez, J.-M.; Peña-Luque, S. Automatic Water Detection from Multidimensional Hierarchical Clustering for Sentinel-2 Images and a Comparison with Level 2A Processors. Remote Sensing of Environment 2021, 253, 112209. https://doi.org/10.1016/j.rse.2020.112209.