Skip to content

Latest commit

 

History

History
533 lines (399 loc) · 21.5 KB

NEWS.md

File metadata and controls

533 lines (399 loc) · 21.5 KB
Raw

SITS Release History

What's new in SITS version 1.5

Hotfix version 1.5.0-1

  • Add multicores processing support for active learning sampling methods
  • Remove tapply from .reg_cube_split_assets() for R 4.X compatibility
  • Fix sits_merge() function that was not merging SAR and OPTICAL cubes
  • Rename n_input_pixels back to input_pixels for compatibility with models trained in old versions of the package
  • Fix torch usage in Apple M3 by turning off MPS technology
  • Fix date parameter usage in sits_view()
  • Improve plot() performance using raster overviews
  • Include support for PLANET Mosaic product

New features in SITS version 1.5.0

  • Support for SENTINEL-1-RTC and SENTINEL-2-L2A in CDSE
  • Include support for DEA products SENTINEL-1-RTC, LS5-SR, LS7-SR, LS9-SR, ALOS-PALSAR-MOSAIC, NDVI ANOMALY, DAILY CHIRPS, MONTHLY CHIRPS and DEM-30
  • Support for Sentinel-1 GRD and RTC collections from Planetary Computer
  • Include parameter tile to select data from Sentinel-1 (MPC)
  • Include parameter tile to select data from Sentinel-1 and Sentinel-2 (DEAFRICA)
  • Include parameter tile to select data from HLS collections
  • Improved support for GPU-based classification of deep learning models
  • Support for non-normalized derived indexes
  • Support for shapefiles as ROI in sits_cube()
  • Fix inconsistencies in HLS scale factors
  • New function to obtain ROI based on MGRS tiles
  • Add support for uncertainty cubes in sits_mosaic()
  • Improve performance of sits_segment() using chunk parallelization
  • Include uncertainty measures for vector probability cubes
  • New sits_clean() function to improve classified maps
  • New functions sits_sampling_design() and sits_stratified_sampling()
  • New sits_reduce() function
  • Include dtw distance when building SOM maps

What's new in SITS version 1.4

Hotfix version 1.4.2-3

  • Fix font download in package initialization

Hotfix version 1.4.2-2

  • Fix integer overflow bug in sits_classify() segments

Hotfix version 1.4.2-1

  • Fix crs bug in sits_apply()
  • Update file name in clean feature
  • Fix time series extraction bug with segments
  • Fix examples

New features in SITS version 1.4.2

  • Support for vector data cubes, including visualisation
  • Object-based time series analysis using spatio-temporal segmentation
  • Improved support for GPU usage when running deep learning algorithms
  • New function to clean values by modal filter in classified images
  • Added experimental support for Sentinel-1 images available on MPC
  • Summary function now includes cloud cover information
  • General bug fixes

New features in SITS version 1.4.1

  • Updated access to collections in Brazil Data Cube, HLS, and AWS
  • Corrected errors in labelling of classified cubes
  • Created a factory of functions for segmentation

New features in SITS version 1.4.0

  • New function for image segmentation based on supercells package
  • New version of sits_get_data() to extract average values of time series based on segments
  • Support for Harmonized Landsat Sentinel (HLS) collections from NASA
  • Support for probability cubes and uncertainty cubes in sits_view()
  • New summary() function to show details of data cubes and time series tibbles
  • General big fixes

What's new in SITS version 1.3

New features in SITS version 1.3.0

  • Remove NOTES and WARNINGS pointed out by CRAN
  • New sits_mosaic() function for improving visualization of large data sets
  • Add support to cubes with no cloud coverage information in sits_regularize()
  • Improve sits_cube_copy() for downloading data from the internet
  • Tested and validated GPU support for deep learning models in sits
  • Added multithread support for deep learning models in sits_train()
  • Improve sits_combine_predictions()
  • Remove dependencies on data.table package
  • Organize and clean internal APIs
  • General bug fixes

What's new in SITS version 1.2.0

Hotfix version 1.2.0-4

  • Fix .raster_file_blocksize.terra() bug (issue #918)

Hotfix version 1.2.0-3

  • Fix stars proxy bug (issue #902)
  • Fix purrr cross deprecation
  • Fix ggplot2 aes_string deprecation

Hotfix version 1.2.0-2

  • Fix tibble subsetting bug (issue #893)

Hotfix version 1.2.0-1

  • Fix sits_som_clean_samples() bug (issue #890)

New features in SITS version 1.2.0

  • sits_get_data() can be used to retrieve samples in classified cube
  • Support for mixture models (sits_mixture_model())
  • Joining cubes in a mosaic (sits_mosaic_cubes())
  • Extract the trained ML model (sits_model())
  • Downloading and copying data cubes (sits_cube_copy())
  • Combine prediction by average and entropy (sits_combine_predictions())
  • Significant performance improvement when working with COG files
  • Allow plot of confusion matrix (sits_plot)
  • Support for operations on CLOUD band in sits_apply()
  • Bug fixes and internal re-engineering for better code maintenance

What's new in SITS version 1.1.0

Hotfix version 1.1.0-8

  • Fix support to BDC cubes in sits_regularize() (issue #848)
  • Fix support to classified_image cubes in sits_labels()<- (issue #846)

Hotfix version 1.1.0-7

  • Fix out of memory error in sits_label_classification() and sits_smooth() (issue #850)

Hotfix version 1.1.0-6

  • Fix resume feature in sits_classify() on BDC cubes (issue #844)

Hotfix version 1.1.0-5

  • Fix bound box issue in image blocks produced by sits_apply()

Hotfix version 1.1.0-4

  • Fix MPC cube time expiration bug

Hotfix version 1.1.0-3

  • Fix bound box issue in image blocks produced by sits_apply()

Hotfix version 1.1.0-2

  • Improve sits_values() function (issue #810)
  • Fix sits_reduce_imbalance() function (issue #809)

Hotfix version 1.1.0-1

  • Fix sits_accuracy() function (issue #807)

New features in SITS version 1.1.0

  • Introduced support to kernel functions in sits_apply
  • Introduced new function sits_mixture_model for spectral mixture analysis
  • Support for the Swiss Data Cube (swissdatacube.org)
  • Support for mosaic visualization in sits_view
  • Introduced new function sits_as_sf to convert sits objects to sf
  • Export images as COG in sits_regularize
  • Add roi parameter in sits_regularize function
  • Add crs parameter in sits_get_data
  • Change Microsoft Planetary Computer source name to "MPC"
  • Fix several bugs and improve performance

What's new in SITS version 1.0.0

  • Available on CRAN.

New features in SITS version 1.0.0

  • Hotfix to improve sits_whittaker() function to process cube.
  • Update documentation to match CRAN standards

What's new in SITS version 0.17.0

New features in SITS version 0.17.0

  • Introduced new classifier model sits_lighttae() (Lightweight Temporal Self-Attention)
  • Introduced sits_uncertainty_sampling() for active learning
  • Introduced sits_confidence_samples() for semi-supervised learning
  • Introduced sits_geo_dist() to generate samples-samples and samples-predicted plot
  • Introduced sits_tuning() for random search of machine learning parameters
  • Introduced sits_reduce_imbalance() function to balance class samples
  • Introduced sits_as_sf() to convert a sits tibble to a sf object
  • Support to torchopt deep learning optimizer package
  • New types of sits_uncertainty(): least confidence and margin of confidence

Improvements in SITS version 0.17.0

  • Implement parallel processing for sits_kfold_validate()
  • Change data to samples in sits machine learning classifiers (NOTE: models trained in previous versions is no longer supported)
  • Change deep learning functions to snake case
  • Remove file parameter in sits_get_data() function
  • Update documentation
  • Improve several internal functions performances
  • Fix several bugs

What's new in SITS version 0.16.3

New features in SITS version 0.16.3

  • reimplemented all deep learning functions using torch package and remove keras dependence
  • Introduced sits_TAE() classification model
  • Introduced sits_lightgbm() classification model
  • Simplified sits_regularize() parameters
  • Improve sits_regularize() to reach production level quality
  • Improve sits_regularize() to use C++ internal functions
  • Include improved version of gdalcubes
  • Improve sits_cube() to open results cube
  • Update plot() parameters on raster cubes
  • Support multi-tile for classified cube in sits_view()

What's new in SITS version 0.16.2

New features in SITS version 0.16.2

  • Improve sits_get_data() to accept tibbles
  • Remove multiples progress bar from sits_cube()
  • Improve sits_regularize() to process in parallel by tiles, bands, and dates
  • Improve sits_regularize() to check malformed files

What's new in SITS version 0.16.1

New features in SITS version 0.16.1

  • Update AWS_NO_SIGN_REQUEST environment variable
  • Solved bug in .gc_get_valid_interval() function.
  • Now sits_regularize has a fault tolerance system, so that if there is a processing error the function will delete the malformed files and create them again.
  • sits_regularize function has a new parameter called multithreads.
  • sits_cube function for local cubes has a new parameter called multicores.
  • Print F1 score in sits_kfold_validate with more than 2 labels.

What's new in SITS version 0.16.0

New features in SITS version 0.16.0-1

  • hotfix sits_cube() function to tolerate malformed paths from STAC service;

New features in SITS version 0.16.0

  • Include sits_apply() function to generate new bands from existing ones;
  • Improve sits_accuracy() function to work with multiple cubes;
  • Add band parameter to sits_view()
  • Introduce sits_uncertainty() function to provide uncertainty measure to probability maps;
  • Improve sits_regularize() by taking least cloud cover by default method to compose images
  • Bug fixes;

What's new in SITS version 0.15.1

New features in SITS version 0.15.1-1

  • Fix bug in sits_regularize that generated images with artifacts
  • Fix wrong bbox in sits_cube from STAC AWS Sentinel-2

New features in SITS version 0.15.1

  • Update README.Rmd
  • Support sits_timeline() to sits model objects
  • Update drone image
  • Simplify config_colors.yml by removing palette names
  • Temporary python files are being generated in the check
  • Organize color handling in SITS
  • Organize configuration files
  • Improve preconditions in sits_regularize()
  • Compress external data with bzip2
  • Update gdalcubes format files
  • Update rstac version
  • Check provided parameters in sits_regularize function
  • Use default palette for SOM colors
  • Remove start_date and end_date from validation csv file
  • Use a default brewer palette to plot classified cube
  • Improve package help pages
  • Remove unused data sets
  • Remove rarely used functions
  • sits_regularize() is producing Float64 images as output
  • Full support for Microsoft Planetary Computing

What's new in SITS version 0.15.0

New features in SITS version 0.15.0-4

  • Change gdalcubes_chunk_size in "config.yml" to improve sits_regularize().

New features in SITS version 0.15.0-3

  • Fix bug in .source_collection_access_test to pass ellipsis to rstac::post_request function.

New features in SITS version 0.15.0-2

  • Fix bug in .source_collection_access_test to pass ellipsis to rstac::post_request function.
  • Update drone version

New features in SITS version 0.15.0-1

  • Fix bug in sits_plot
  • Fix bug in sits_timeline for cubes that do not have the same temporal extent.

New features in SITS version 0.15.0

  • Support for regularization of collections in DEAFRICA and USGS improvement
  • Collection S2_10_16D_STK-1 removed from BDC source in config file
  • Add a color for NoClass label improvement
  • Change mapview to leaflet package
  • Standardize cube creation parameters
  • Remove CLASSIFIED and PROBS sources from config file
  • Change minimal version requirement of terra package to 1.4-11
  • Update sits_list_collections() to indicate open data collection
  • Geographical visualization of samples
  • Remove dependencies on packages ptw, signal and MASS
  • Add support to open_data collections in config file
  • Change default output_dir parameter
  • Remove sits_cube_clone() function
  • Plot RGB images from raster cubes
  • Fixed error in sits_select() for bands in raster cube
  • Update examples in demo
  • Support open data collections of DEAFRICA and AWS
  • Support USGS STAC Landsat 8 catalog
  • User can provide resampling method to sits_regularize() function
  • Add support to open data collections on 'AWS' source
  • Remove OPENDATA source
  • Update documentation
  • Resolve ambiguity in "bands" parameter for data cubes
  • Remove "sits_bands" assignment function
  • Include "labels" information only on probs and labelled data cubes
  • Remove S2_10-1 BDC collection from config
  • Other bug fixes

What's new in SITS version 0.14.1

New features in SITS version 0.14.1-1

  • Bug in cube generated by sits_regularize() cannot have "CLOUD" band

New features in SITS version 0.14.1

  • Implement new function sits_list_collections()
  • Update gdalcubes parameters
  • Implement .source_bands_resampling()
  • Remove name from demo file
  • Improve sits_som_clean_samples() function
  • Improve sits_bands<-() function
  • Improve sits_select() function
  • Error in cloud band of CBERS4 data example
  • Include a function to list collections available in cloud services
  • sits_cube_copy() does not include information on the tile
  • Get spatial resolution from config file
  • Fix partial merge configuration file
  • Change bbox to roi in sits

What's new in SITS version 0.14.0

New features in SITS version 0.14.0-2

  • fix sits_bbox() function

New features in SITS version 0.14.0-1

  • fix duplicate link in AWS STAC

New features in SITS version 0.14.0

  • Now the plot of a classified cube requires a legend or a palette if the labels are not in the default sits palette.
  • Support for S2-SEN2COR_10_16D_STK-1 BDC collection
  • Remove function name from msg in check function
  • Add satellite and sensor info in config file
  • Remove imager, ranger, proto, and future packages from sits
  • Support for different providers to LOCAL sources
  • LOCAL source is dynamically built
  • Remove sits_cube.local_cube() function parameters satellite and sensor
  • Add parameters origin and collection to sits_cube.local_cube() function
  • Fix LOCAL source examples and tests

What's new in SITS version 0.13.1

New features in SITS version 0.13.1

  • Update and add more tests in CI
  • Implement new check functions
  • Change error and warning messages
  • fix deprecated warnings in keras package
  • bug fixes

What's new in SITS version 0.13.0-3

New features in SITS version 0.13.0-3

  • Update documentation in Machine Learning methods
  • Hotfix bug in neuron labelling

What's new in SITS version 0.13.0-2

New features in SITS version 0.13.0-2

  • Bug fixes in BDC MODIS cube

What's new in SITS version 0.13.0-1

New features in SITS version 0.13.0-1

  • Bug fixes in check STAC bands
  • Change Landsat-8 (LC8_30-1) product metadata for BDC source

What's new in SITS version 0.13

New features in SITS version 0.13.0

  • Create API for source cube
  • Update auxiliary functions of the config file
  • Update config file
  • Add support to others bands values in config file
  • Add support to bit mask in USGS cube
  • Support to multiples directories in local cubes
  • Support for MODIS cloud bands
  • Dealing with invalid areas in SITS
  • Support for WTSS

What's new in SITS version 0.12

New features in SITS version 0.12.1

  • Update README
  • Change docker image to new sits build
  • Adjust CMASK bands values in BDC cubes
  • Support for sits_config_sensor_bands accept more than one sensor
  • sits cube selection by shapefile
  • Problem - sits classify

New features in SITS version 0.12.0

  • Bugs fixed
  • Documentation updated
  • Support for multiple tile in local cubes
  • Improve selection using roi parameter in sits_classify() function

What's new in SITS version 0.11

New features in SITS version 0.11.2

  • Added keras serialisation to TempCNN and ResNet models

New features in SITS version 0.11.1

  • Removed LSTM and FCN deep learning models

New features in SITS version 0.11.0

  • Important improvements in classification performance
  • Updated version of deep learning methods
  • Support for STAC access to Brazil Data Cube, AWS and DE Africa
  • Improved sits validation

What's new in SITS version 0.10

New features in SITS version 0.10.0

  • Version update 0.10.0
  • Continuous Integration (drone.io)
  • Bayesian smoothing improvement
  • Introduces Snow multiprocessing architecture
  • cube plot allow region of interest (roi)
  • Support for multiple tiles
  • Update documentation
  • Bugs fix

What's new in SITS version 0.9

New features in SITS version 0.9.8

  • Access to Sentinel-2 level-2A images in AWS
  • Access to the Brazil Data Cube using STAC
  • Improved raster API
  • Code revision with lintr and good practices packages
  • Improvement of assertions and code coverage
  • Examples and tests generate output in tempdir()

New features in SITS version 0.9.7

  • Image classification using region of interest (ROI)

New features in SITS version 0.9.6

  • Access and processing of tiles of the Brazil Data Cube
  • Plotting of data cube and probability images
  • Examples of using SITS with SENTINEL-2 and CBERS-4 images

New features in SITS version 0.9.5

  • Time series tibbles and data cube metadata can now be saved and read in SQLite
  • Code coverage increased to 95%
  • Vignettes have been moved to "sits-docs" to reduce building time

New features in SITS version 0.9.4

  • Filtering can be applied to classified images
  • Band suffix in filtering is now set to ""
  • Improvement in code coverage: most of the code has more than 90% coverage

New features in SITS version 0.9.3

  • Improvements in reading shapefiles: using sampling to retrieve time series inside polygons
  • Improvement is plotting: uses overloading to the "plot" function

New features in SITS version 0.9.2

  • Raster classification results can now have versions: a new parameter "version" has been included in the sits_classify function.

  • Corrections to sits_kohonen and to the documentation.

New Features in version 0.9.1

  • New deep learning models for time series: 1D convolutional neural networks (sits_FCN), combining 1D CNN and multi-layer perceptron networks (sits_TempCNN), 1D version of ResNet (sits_ResNet), and combination of long-short term memory (LSTM) and 1D CNN (sits_LSTM_FCN).

  • New version of area accuracy measures that include Olofsson metrics ()

What's new in SITS version 0.8

New Features

  • From version 0.8 onwards, the package has been designed to work with data cubes. All references to "coverage" have been replaced by references to "cubes".

  • The classification of raster images using sits_classify now produces images with the information on the probability of each class for each pixel. This allows more flexibility in the options for labeling the resulting probability raster files.

  • The function sits_label_classification has been introduced to generate a labelled image from the class probability files, with optional smoothing. The choices are smoothing = none (default), smoothing = bayesian (for bayesian smoothing) and smoothing = majority (for majority smoothing).

  • To better define a cube, the metadata tibble associated to a cube requires four parameters to define the cube: (a) the web service that provides time series or cubes; (b) the URL of the web service; (c) the name of the satellite; (d) the name of the satellite sensor. If not provided, these parameters are inferred for the sits configuration file.

  • The functions that do data transformations, such as sits_tasseled_cap and sits_savi now require a sensor parameter ("MODIS" is the default)

  • Functions sits_bands and sits_labels now work for both tibbles with time series and data cubes.

Configuration file

  • The SITS configuration file has been improved to include information about web service providers, satellites and sensor parameters. Please use sits_show_config() to see the default contents. Users can override these parameters or add their own by creating a config.yml file in their home directory.

Examples and Demos

  • Examples and demos that include classification of raster files now use the inSitu R package, available using devtools::install_github(e-sensing/inSitu).

  • All examples have been tested and checked for correctness.

Functions removed

  • sits_coverage has been replaced by sits_cube.

  • sits_raster_classification has been removed. Please use sits_classify.

  • In sits_classify, the parameter out_prefix has been changed to output_dir, to allow better control of the directory on which to write.

  • sits_bayes_smooth has been removed. Please use sits_label_classification with smoothing = bayesian.

  • To define a cube based on local files, service = RASTER has been replaced by service = LOCALHOST.

Improvements and fixes

  • For programmers only: The sits_cube.R file now includes many convenience functions to avoid using cumbersome indexes to files and vector: .sits_raster_params, .sits_cube_all_robjs, .sits_class_band_name, .sits_cube_bands, .sits_cube_service, .sits_cube_file, .sits_cube_files, .sits_cube_labels, .sits_cube_timeline, .sits_cube_robj, .sits_cube_all_robjs, .sits_cube_missing_values, .sits_cube_minimum_values, .sits_cube_maximum_values, .sits_cube_scale_factors, .sits_files_robj. Please look at the documentation provided in the sits_cube.R file.

  • For programmers only: The metadata that describes the data cube no longer stores the raster objects associated to the files associated with the cube.