Satellite Image Time Series Analysis for Earth Observation Data Cubes
Cross-validate time series samples
Print the values of a confusion matrix
Print the area-weighted accuracy
Print accuracy summary
Check is date is valid
histogram of prob cubes
histogram of data cubes
Histogram
Histogram uncertainty cubes
Replace NA values with linear interpolation
Plot classified images
Plot Segments
Plot DEM cubes
Make a kernel density plot of samples distances.
Plot patterns that describe classes
Plot time series predictions
Assess classification accuracy (area-weighted method)
Plot probability cubes
Plot probability vector cubes
Plot RGB data cubes
Plot time series
Plot Random Forest model
Function to show colors in SITS
Plot SAR data cubes
Plot confusion matrix
Plot a dendrogram cluster
Plot confusion between clusters
Plot a SOM map
Plot Torch (deep learning) model
Function to retrieve sits color table
Plot uncertainty cubes
Plot uncertainty vector cubes
Plot variance cubes
Plot RGB vector data cubes
Plot XGB model
Add base maps to a time series data cube
Apply a function on a set of time series
Return a sits_tibble or raster_cube as an sf object.
Get the names of the bands
Get the bounding box of the data
Classify time series or data cubes
Create a closure for calling functions with and without data
Cleans a classified map using a local window
Removes labels that are minority in each cluster.
Find clusters in time series samples
Show label frequency in each cluster produced by dendrogram analysis
Function to save color table as QML style for data cube
Function to reset sits color table
Function to set sits color table
Train multi-layer perceptron models using torch
Estimate ensemble prediction based on list of probs cubes
Suggest high confidence samples to increase the training set.
Show current sits configuration
List the cloud collections supported by sits
Configure parameters for sits package
Copy the images of a cube to a local directory
Create data cubes from image collections
Filter time series with smoothing filter
Define a linear formula for classification models
Define a loglinear formula for classification models
Compute the minimum distances among samples and prediction points.
Get time series from data cubes and cloud services
Replace NA values in time series with imputation function
Build a labelled image from a probability cube
Inform label distribution of a set of time series
Get labels associated to a data set
Train a model using Lightweight Temporal Self-Attention Encoder
List the cloud collections supported by sits
Merge two data sets (time series or cubes)
Convert MGRS tile information to ROI in WGS84
Multiple endmember spectral mixture analysis
Export classification models
Mosaic classified cubes
Find temporal patterns associated to a set of time series
Obtain numerical values of predictors for time series samples
Normalize predictor values
Obtain categorical id and predictor labels for time series samples
Obtain a fraction of the predictors data frame
Obtain predictors for time series samples
Reclassify a classified cube
Reduce imbalance in a set of samples
Reduces a cube or samples from a summarization function
Build a regular data cube from an irregular one
Train random forest models
Informs if sits examples should run
Informs if sits tests should run
Sample a percentage of a time series
Allocation of sample size to strata
Segment an image
Filter bands on a data set (tibble or cube)
Filter time series with Savitzky-Golay filter
Shows the predicted labels for a classified tibble
Segment an image using SLIC
Smooth probability cubes with spatial predictors
Cleans the samples based on SOM map information
Evaluate cluster
Evaluate cluster
Use SOM for quality analysis of time series samples
Obtain statistics for all sample bands
Allocation of sample size to strata
Train support vector machine models
Train a model using Temporal Self-Attention Encoder
Train temporal convolutional neural network models
Get timeline of a cube or a set of time series
Export a sits tibble metadata to the CSV format
Save accuracy assessments as Excel files
Train classification models
Tuning machine learning models hyper-parameters
Tuning machine learning models hyper-parameters
Suggest samples for enhancing classification accuracy
Estimate classification uncertainty based on probs cube
Validate time series samples
Calculate the variance of a probability cube
View data cubes and samples in leaflet
Filter time series with whittaker filter
Train extreme gradient boosting models
sits
Summarize data cubes
Summarize data cubes
Summarize accuracy matrix for training data
Summarize accuracy matrix for area data
Summarize sits
Summarise variance cubes
Change the labels of a set of time series
An end-to-end toolkit for land use and land cover classification using big Earth observation data, based on machine learning methods applied to satellite image data cubes, as described in Simoes et al (2021) <doi:10.3390/rs13132428>. Builds regular data cubes from collections in AWS, Microsoft Planetary Computer, Brazil Data Cube, Copernicus Data Space Environment (CDSE), Digital Earth Africa, Digital Earth Australia, NASA HLS using the Spatio-temporal Asset Catalog (STAC) protocol (<https://stacspec.org/>) and the 'gdalcubes' R package developed by Appel and Pebesma (2019) <doi:10.3390/data4030092>. Supports visualization methods for images and time series and smoothing filters for dealing with noisy time series. Includes functions for quality assessment of training samples using self-organized maps as presented by Santos et al (2021) <doi:10.1016/j.isprsjprs.2021.04.014>. Provides machine learning methods including support vector machines, random forests, extreme gradient boosting, multi-layer perceptrons, temporal convolutional neural networks proposed by Pelletier et al (2019) <doi:10.3390/rs11050523>, and temporal attention encoders by Garnot and Landrieu (2020) <doi:10.48550/arXiv.2007.00586>. Supports GPU processing of deep learning models using torch <https://torch.mlverse.org/>. Performs efficient classification of big Earth observation data cubes and includes functions for post-classification smoothing based on Bayesian inference, and methods for active learning and uncertainty assessment. Supports object-based time series analysis using package supercells <https://jakubnowosad.com/supercells/>. Enables best practices for estimating area and assessing accuracy of land change as recommended by Olofsson et al (2014) <doi:10.1016/j.rse.2014.02.015>. Minimum recommended requirements: 16 GB RAM and 4 CPU dual-core.
Useful links