Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/e-sensing/sits
Satellite image time series in R
https://github.com/e-sensing/sits
big-earth-data cbers earth-observation eo-datacubes geospatial image-time-series land-cover-classification landsat planetary-computer r-spatial remote-sensing rspatial satellite-image-time-series satellite-imagery sentinel-2 stac-api stac-catalog
Last synced: 3 months ago
JSON representation
Satellite image time series in R
- Host: GitHub
- URL: https://github.com/e-sensing/sits
- Owner: e-sensing
- License: gpl-2.0
- Created: 2017-07-27T13:32:40.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2024-08-02T18:14:09.000Z (3 months ago)
- Last Synced: 2024-08-03T21:34:28.337Z (3 months ago)
- Topics: big-earth-data, cbers, earth-observation, eo-datacubes, geospatial, image-time-series, land-cover-classification, landsat, planetary-computer, r-spatial, remote-sensing, rspatial, satellite-image-time-series, satellite-imagery, sentinel-2, stac-api, stac-catalog
- Language: R
- Homepage: https://e-sensing.github.io/sitsbook/
- Size: 275 MB
- Stars: 450
- Watchers: 30
- Forks: 76
- Open Issues: 9
-
Metadata Files:
- Readme: README.Rmd
- Changelog: NEWS.md
- Contributing: contributing.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Codemeta: codemeta.json
Awesome Lists containing this project
README
---
title: "SITS - Satellite Image Time Series Analysis for Earth Observation Data Cubes"
authors: Rolf Simoes, Gilberto Camara, Felipe Souza, Lorena Santos, Pedro Andrade, Alexandre Carvalho, Gilberto Queiroz, Karine Ferreira
output: github_document
---
```{r, echo=FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-"
)
# set seed for reproducibility
set.seed(4242)
torch::torch_manual_seed(1234)
```
[](https://github.com/ropensci/software-review/issues/596)
[](https://cran.r-project.org/package=sits)
[](https://github.com/e-sensing/sits/actions/workflows/R-CMD-check.yaml)
[](https://codecov.io/gh/e-sensing/sits)
[](https://e-sensing.github.io/sitsbook/)
[](https://lifecycle.r-lib.org/articles/stages.html)
[](https://github.com/e-sensing/sits/blob/master/LICENSE)
## Overview
`sits` is an open source R package for satellite image time series analysis. It enables users to apply machine learning techniques for classifying image time series obtained from earth observation data cubes. The basic workflow in `sits` is:
1. Select an image collection available on cloud providers AWS, Microsoft Planetary Computer, Digital Earth Africa, Swiss Data Cube, NASA Harmonized Landsat/Sentinel and Brazil Data Cube.
2. Build a regular data cube from analysis-ready image collections.
3. Extract labelled time series from data cubes to be used as training samples.
4. Perform samples quality control using self-organised maps.
5. Train machine learning and deep learning models.
6. Tune deep learning models for improved accuracy.
7. Classify data cubes using machine learning and deep learning models.
8. Run spatial-temporal segmentation methods for object-based time series classification.
9. Post-process classified images with Bayesian smoothing to remove outliers.
10. Estimate uncertainty values of classified images.
11. Evaluate classification accuracy using best practices.
12. Improve results with active learning and self-supervised learning methods.
```{r sits, echo=FALSE, out.width="60%", out.height="60%", fig.align="center", fig.cap="Conceptual view of data cubes (source: authors)"}
knitr::include_graphics("inst/extdata/markdown/figures/sits_general_view.jpg")
```
## Documentation
Detailed documentation on how to use `sits` is available in the e-book ["Satellite Image Time Series Analysis on Earth Observation Data Cubes"](https://e-sensing.github.io/sitsbook/).
## `sits` on Kaggle
Those that want to evaluate the `sits` package before installing are invited to run the examples available on [Kaggle](https://www.kaggle.com/esensing/code). If you are new on kaggle, please follow the [instructions](https://gist.github.com/OldLipe/814089cc5792c9c0c989d870a22910f4) to set up your account. These examples provide a fast-track introduction to the package. We recommend running them in the following order:
1. [Introduction to SITS](https://www.kaggle.com/esensing/introduction-to-sits)
2. [Working with time series in SITS](https://www.kaggle.com/esensing/working-with-time-series-in-sits)
3. [Creating data cubes in SITS](https://www.kaggle.com/esensing/creating-data-cubes-in-sits)
4. [Improving the quality of training samples](https://www.kaggle.com/code/esensing/improving-quality-of-training-samples)
5. [Machine learning for data cubes](https://www.kaggle.com/esensing/machine-learning-for-data-cubes)
6. [Classification of raster data cubes](https://www.kaggle.com/code/esensing/classification-of-raster-data-cubes)
7. [Bayesian smoothing for post-processing](https://www.kaggle.com/code/esensing/bayesian-smoothing-for-post-processing)
8. [Uncertainty and active learning](https://www.kaggle.com/code/esensing/uncertainty-and-active-learning)
9. [Object-based time series classification](https://www.kaggle.com/esensing/object-based-image-time-series-classification)
## Installation
### Pre-Requisites
The `sits` package relies on the geospatial packages `sf`, `stars`, `gdalcubes` and `terra`, which depend on the external libraries GDAL and PROJ. Please follow the instructions for installing `sits` from the [Setup chapter of the on-line sits book](https://e-sensing.github.io/sitsbook/setup.html).
### Obtaining `sits`
`sits` can be installed from CRAN:
```{r, eval=FALSE}
install.packages("sits")
```
The latest supported version is available on github. It may have additional fixes from the version available from CRAN.
```{r, eval=FALSE}
devtools::install_github("e-sensing/sits", dependencies = TRUE)
```
```{r}
# load the sits library
library(sits)
```
### Support for GPU
Classification using torch-based deep learning models in `sits` uses CUDA compatible NVIDIA GPUs if available, which provides up 10-fold speed-up compared to using CPUs only. Please see the [installation instructions](https://torch.mlverse.org/docs/articles/installation) for more information on how to install the required drivers.
## Building Earth Observation Data Cubes
### Image Collections Accessible by `sits`
Users create data cubes from analysis-ready data (ARD) image collections available in cloud services. The collections accessible in `sits` `r packageVersion("sits")` are:
1. Brazil Data Cube ([BDC](http://brazildatacube.org/en/home-page-2/#dataproducts)): Open data collections of Sentinel-2, Landsat-8 and CBERS-4 images.
2. Microsoft Planetary Computer ([MPC](https://planetarycomputer.microsoft.com/catalog)): Open data collection of Sentinel-2/2A and Landsat-8
3. Earth on AWS ([AWS](https://aws.amazon.com/earth/)): Sentinel-2/2A level 2A collections.
4. Digital Earth Africa ([DEAFRICA](https://www.digitalearthafrica.org/)): Open data collection of Sentinel-2/2A and Landsat-8 for Africa.
5. [USGS](https://landsatlook.usgs.gov/stac-browser): Landsat-4/5/7/8 collections, which are not open data.
6. Swiss Data Cube ([SDC](https://www.swissdatacube.org/)): Open data collection of Sentinel-2/2A and Landsat-8.
7. NASA Harmonized Landsat/Sentinel Collection [HLS](https://hls.gsfc.nasa.gov/).
Open data collections do not require payment of access fees. Except for those in the Brazil Data Cube, these collections are not regular. Irregular collections require further processing before they can be used for classification using machine learning models.
### Building a Data Cube from an ARD Image Collection
The following code defines an irregular data cube of Sentinel-2/2A images available in the Microsoft Planetary Computer, using the open data collection `"SENTINEL-2-L2A"`. The geographical area of the data cube is defined by the tiles `"20LKP"` and `"20LLKP"`, and the temporal extent by a start and end date. Access to other cloud services works in similar ways.
```{r, tidy = "styler", echo=TRUE}
s2_cube <- sits_cube(
source = "MPC",
collection = "SENTINEL-2-L2A",
tiles = c("20LKP", "20LLP"),
bands = c("B03", "B08", "B11", "SCL"),
start_date = as.Date("2018-07-01"),
end_date = as.Date("2019-06-30"),
progress = FALSE
)
```
This cube is irregular. The timelines of tiles `"20LKP"` and `"20LLKP"` and the resolutions of the bands are different. Sentinel-2 bands `"B03"` and `"B08"` have 10-meters resolution, while band `"B11"` and the cloud band `"SCL"` have 20-meters resolution. Irregular collections need an additional processing step to be converted to regular data cubes, as described below.
```{r dcconcept, echo=FALSE, out.width="90%", out.height="90%", fig.align="center", fig.cap="Conceptual view of data cubes (source: authors)"}
knitr::include_graphics("inst/extdata/markdown/figures/datacube_conception.jpg")
```
After defining an irregular ARD image collection from a cloud service using `sits_cube()`, users should run `sits_regularize()` to build a regular data cube. This function uses the [gdalcubes R package](https://github.com/appelmar/gdalcubes), described in [Appel and Pebesma, 2019](https://www.mdpi.com/2306-5729/4/3/92).
```{r, tidy = "styler", eval=FALSE, echo=TRUE}
gc_cube <- sits_regularize(
cube = s2_cube,
output_dir = tempdir(),
period = "P15D",
res = 60,
multicores = 4
)
```
The above command builds a regular data cube with all bands interpolated to 60 m spatial resolution and 15-days temporal resolution. Regular data cubes are the input to the `sits` functions for time series retrieval, building machine learning models, and classification of raster images and time series.
The cube can be shown in a leaflet using `sits_view()`.
```{r, echo=TRUE, eval=FALSE}
# View a color composite on a leaflet
sits_view(s2_cube[1, ], green = "B08", blue = "B03", red = "B11")
```
## Working with Time Series in `sits`
### Accessing Time Series in Data Cubes
`sits` has been designed to use satellite image time series to derive machine learning models. After the data cube has been created, time series can be retrieved individually or by using CSV or SHP files, as in the following example. The example below uses a data cube in a local directory, whose images have been obtained from the `"MOD13Q1-6"` collection of the Brazil Data Cube.
```{r, tidy = "styler", eval=TRUE, echo=TRUE, cache=TRUE}
library(sits)
# this data cube uses images from the Brazil Data Cube that have
# downloaded to a local directory
data_dir <- system.file("extdata/raster/mod13q1", package = "sits")
# create a cube from downloaded files
raster_cube <- sits_cube(
source = "BDC",
collection = "MOD13Q1-6",
data_dir = data_dir,
delim = "_",
parse_info = c("X1", "X2", "tile", "band", "date"),
progress = FALSE
)
# obtain a set of samples defined by a CSV file
csv_file <- system.file("extdata/samples/samples_sinop_crop.csv",
package = "sits"
)
# retrieve the time series associated with the samples from the data cube
points <- sits_get_data(raster_cube, samples = csv_file)
# show the time series
points[1:3, ]
```
After a time series has been obtained, it is loaded in a tibble. The first six columns contain the metadata: spatial and temporal location, label assigned to the sample, and coverage from where the data has been extracted. The spatial location is given in longitude and latitude coordinates. The first sample has been labelled "Pasture", at location (-55.65931, -11.76267), and is considered valid for the period (2013-09-14, 2014-08-29).
## Time Series Classification
### Training Machine Learning Models
`sits` provides support for the classification of both individual time
series as well as data cubes. The following machine learning methods are
available in `sits`:
- Support vector machines (`sits_svm()`)
- Random forests (`sits_rfor()`)
- Extreme gradient boosting (`sits_xgboost()`)
- Multi-layer perceptrons (`sits_mlp()`)
- Deep Residual Networks (`sits_resnet()`) (see ref. [8])
- 1D convolution neural networks (`sits_tempcnn()`) (see ref. [9])
- Temporal self-attention encoder (`sits_tae()`) (see ref. [10])
- Lightweight temporal attention encoder (`sits_lighttae()`) (see ref. [11] and [12])
The following example illustrate how to train a dataset and classify an
individual time series. First we use the `sits_train()` function with two
parameters: the training dataset (described above) and the chosen
machine learning model (in this case, TempCNN).
The trained model is then used to classify a time series from Mato Grosso
Brazilian state, using `sits_classify()`. The results can be shown in text
format using the function `sits_show_prediction()` or graphically using
`plot`.
```{r, tidy = "styler", fig.align="center", fig.height=3.5, fig.width=7, fig.cap="Classification of NDVI time series using TempCNN", cache=TRUE}
# training data set
data("samples_modis_ndvi")
# point to be classified
data("point_mt_6bands")
# Train a deep learning model
tempcnn_model <- sits_train(
samples = samples_modis_ndvi,
ml_method = sits_tempcnn()
)
# Select NDVI band of the point to be classified
# Classify using TempCNN model
# Plot the result
point_mt_6bands |>
sits_select(bands = "NDVI") |>
sits_classify(tempcnn_model) |>
plot()
```
The following example shows how to classify a data cube organized as a
set of raster images. The result can also be visualized interactively
using `sits_view()`.
```{r, tidy = "styler", fig.align="center", fig.height=4.5, fig.width=9, fig.cap="Land use and Land cover in Sinop, MT, Brazil in 2018", cache=TRUE}
# Create a data cube to be classified
# Cube is composed of MOD13Q1 images from the Sinop region in Mato Grosso (Brazil)
data_dir <- system.file("extdata/raster/mod13q1", package = "sits")
sinop <- sits_cube(
source = "BDC",
collection = "MOD13Q1-6",
data_dir = data_dir,
delim = "_",
parse_info = c("X1", "X2", "tile", "band", "date"),
progress = FALSE
)
# Classify the raster cube, generating a probability file
# Filter the pixels in the cube to remove noise
probs_cube <- sits_classify(
data = sinop,
ml_model = tempcnn_model,
output_dir = tempdir()
)
# apply a bayesian smoothing to remove outliers
bayes_cube <- sits_smooth(
cube = probs_cube,
output_dir = tempdir()
)
# generate a thematic map
label_cube <- sits_label_classification(
cube = bayes_cube,
output_dir = tempdir()
)
# plot the the labelled cube
plot(label_cube,
title = "Land use and Land cover in Sinop, MT, Brazil in 2018"
)
```
## Additional information
Since version 1.4.2, `sits` support OBIA analysis of image time series, using an extension of R package `supercells`.
The package is described in detail in on-line book ["SITS: Data analysis and machine learning for data cubes using satellite image time series"](https://e-sensing.github.io/sitsbook/).
### References
#### Citable papers for sits
If you use `sits`, please cite the following paper:
- [1] Rolf Simoes, Gilberto Camara, Gilberto Queiroz, Felipe Souza, Pedro R. Andrade, Lorena Santos, Alexandre Carvalho, and Karine Ferreira. “Satellite Image Time Series Analysis for Big Earth Observation Data”. Remote Sensing, 13: 2428, 2021. .
Additionally, the sample quality control methods that use self-organized maps are described in the following reference:
- [2] Lorena Santos, Karine Ferreira, Gilberto Camara, Michelle Picoli, Rolf Simoes, “Quality control and class noise reduction of satellite image time series”. ISPRS Journal of Photogrammetry and Remote Sensing, 177:75-88, 2021. .
#### Papers that use sits to produce LUCC maps
- [3] Rolf Simoes, Michelle Picoli, et al., "Land use and cover maps for Mato Grosso State in Brazil from 2001 to 2017". Sci Data 7(34), 2020. .
- [4] Michelle Picoli, Gilberto Camara, et al., “Big Earth Observation Time Series Analysis for Monitoring Brazilian Agriculture”. ISPRS Journal of Photogrammetry and Remote Sensing, 2018. .
- [5] Karine Ferreira, Gilberto Queiroz et al., "Earth Observation Data Cubes for Brazil: Requirements, Methodology and Products". Remote Sens. 12:4033, 2020. .
#### Papers that describe software used in sits
We thank the authors of these papers for making their code available to be used in connection with sits.
- [6] Marius Appel and Edzer Pebesma, “On-Demand Processing of Data Cubes from Satellite Image Collections with the Gdalcubes Library.” Data 4 (3): 1–16, 2020. .
- [7] Ron Wehrens and Johannes Kruisselbrink, "Flexible Self-Organising Maps in kohonen 3.0". Journal of Statistical Software, 87(7), 2018. .
- [8] Hassan Fawaz, Germain Forestier, Jonathan Weber, Lhassane Idoumghar, and Pierre-Alain Muller, "Deep learning for time series classification: a review". Data Mining and Knowledge Discovery, 33(4): 917--963, 2019. .
- [9] Charlotte Pelletier, Geoffrey I. Webb, and Francois Petitjean. “Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series.” Remote Sensing 11 (5), 2019. .
- [10] Vivien Garnot, Loic Landrieu, Sebastien Giordano, and Nesrine Chehata, "Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attention", Conference on Computer Vision and Pattern Recognition, 2020. .
- [11] Vivien Garnot, Loic Landrieu, "Lightweight Temporal Self-Attention for Classifying Satellite Images Time Series", 2020. .
- [12] Maja Schneider, Marco Körner, "[Re] Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attention." ReScience C 7 (2), 2021. .
- [13] Jakub Nowosad, Tomasz Stepinski, "Extended SLIC superpixels algorithm for applications to non-imagery geospatial rasters". International Journal of Applied Earth Observation and Geoinformation, 112, 102935, 2022.
- [14] Martin Tennekes, “tmap: Thematic Maps in R.” Journal of Statistical Software, 84(6), 1–39, 2018.
### Acknowledgements for community support
The authors are thankful for the contributions of Edzer Pebesma, Jakub Novosad. Marius Appel, Martin Tennekes, Robert Hijmans, Ron Wehrens, and Tim Appelhans, respectively chief developers of the packages `sf`/`stars`, `supercells`, `gdalcubes`, `tmap`, `terra`, `kohonen`, and `leafem`. The `sits` package is also much indebted to the work of the RStudio team, including the `tidyverse`. We are indepted to Daniel Falbel for his great work in the `torch` and `luz` packages. We thank Charlotte Pelletier and Hassan Fawaz for sharing the python code that has been reused for the TempCNN and ResNet machine learning models. We would like to thank Maja Schneider for sharing the python code that helped the implementation of the `sits_lighttae()` and `sits_tae()` model. We recognise the importance of the work by Chris Holmes and Mattias Mohr on the STAC specification and API.
### Acknowledgements for Financial and Material Support
We acknowledge and thank the project funders that provided financial and material support:
1. Amazon Fund, established by the Brazilian government with financial contribution from Norway, through the project contract between the Brazilian Development Bank (BNDES) and the Foundation for Science, Technology and Space Applications (FUNCATE), for the establishment of the Brazil Data Cube, process 17.2.0536.1.
2. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brasil (CAPES) and from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), for providing MSc and PhD scholarships.
3. Sao Paulo Research Foundation (FAPESP) under eScience Program grant 2014/08398-6, for for providing MSc, PhD and post-doc scholarships, equipment, and travel support.
4. International Climate Initiative of the Germany Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safety (IKI) under grant 17-III-084- Global-A-RESTORE+ (“RESTORE+: Addressing Landscape Restoration on Degraded Land in Indonesia and Brazil”).
5. Microsoft Planetary Computer under the GEO-Microsoft Cloud Computer Grants Programme.
6. The Open-Earth-Monitor Cyberinfratructure project, which has received
funding from the European Union's Horizon Europe research and innovation programme
under [grant agreement No. 101059548](https://cordis.europa.eu/project/id/101059548).
### How to contribute
The `sits` project is released with a [Contributor Code of Conduct](https://github.com/e-sensing/sits/blob/master/CODE_OF_CONDUCT.md).
By contributing to this project, you agree to abide by its terms.