Skip to content

Commit

Permalink
v24.12
Browse files Browse the repository at this point in the history
  • Loading branch information
XavierCLL committed Dec 21, 2024
1 parent 5992916 commit c641f7a
Showing 1 changed file with 22 additions and 50 deletions.
72 changes: 22 additions & 50 deletions metadata.txt
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@ name=AcATaMa
description=AcATaMa is a powerful and complete Qgis plugin for accuracy assessment that provides comprehensive support for sampling, response design and estimation.
about=AcATaMa is a powerful and complete Qgis plugin for accuracy assessment. It provides comprehensive support for sampling, response design and estimation within a design-based inference framework. The primary goal of AcATaMa is to equip users with the necessary tools to comply with international guidance and best practices for sampling design, estimation of land category areas and changes, and map accuracy assessment. The plugin supports accuracy assessment for any thematic map across diverse use cases. A common application is in remote sensing, particularly for land use and land cover classification, environmental monitoring, urban planning, forestry, agriculture, and more. However, AcATaMa is not limited to these applications; some modules, such as sampling and labeling, can be used independently for other user cases.

version=24.10
version=24.12
qgisMinimumVersion=3.30
qgisMaximumVersion=3.99

Expand All @@ -15,55 +15,27 @@ tracker=https://github.com/SMByC/AcATaMa/issues
repository=https://github.com/SMByC/AcATaMa

changelog=
24.10:
- Major UI refactoring, simplifying and improving usability in the main dockwidget and converting the sampling design section as a window. Many other changes.
- Many changes to adjust and fix the sampling map in stratified sampling. Print the sampling report when the sampling map and thematic map are different. Other fixes.
- Added the nodata value option for the post-stratification map in the simple and systematic sampling design
- A sampling report has been added, containing all the information about the generated samples, their distribution on thematic maps, and other relevant details. This report helps the user determine whether the generated samples meet their expectations before proceeding to the response design process.
- Enable reload sampling file button only when there is a valid sampling file selected
- Some bug fixes to support older versions of the Atacama yaml configuration
- Update and increase the svg internal size to improve the drawing resolution, other UI adjustments
- If sampling change has changed, and user reload it, report is not longer valid
- Ensure that geometry is ok when the sampling file has changed
- Improving the tooltip for the user's accuracy confidence
- Restore response design window when loading a yaml file
- AcATaMa now will save relative paths to the YAML file if the layers are in the same directory or subdirectory tree as the YAML file, making it a portable and shareable project.
- Update the band and nodata from the thematic to sampling design window
- Added more warning notes in the sampling report and adjustments in other sampling report items
- Adding a warning box about the minimum samples in strata for stratified and post-stratified sampling
- Added a new option in stratified sampling to adjust the number of samples based on minimum samples per stratum
- Check and disable the classes with zero pixels on the map in the stratified sampling table
- Adjustments for the standard error of the estimated overall accuracy and User's accuracy confidence default values based on Olofsson et al. (2014)
- Check first if the QtWebKit is available in QT5 client to avoid showing an error every time the user opening the response design window
- Clean the sampling design window before restore the plugin
- While a state of Acatama is restoring, close all the window instances of the plugin opened before restore configuration
- Added some missing connections in the sampling design window
- Convert np.NaN to np.nan for numpy > 2.0
- Replace the variable of the sampling design instance
- Move the references to another page and improving the sampling unit documentation
- Added icons and improving the information text for each sampling type
- All the warning and information messages bar in the sampling design is showing in the window and not in the generic Qgis message bar
- Added stratified method for stratified sampling report
- Use QgsUnitTypes for manage the area unit
- Some improvements for the clear and reload function
- Improve the reload and zoom buttons in the accuracy assessment window
- Fix in counting num samples
- Fixed and issue computing the analysis when pixel value to count is nodata in thematic map
- Fixbug when processing the sampling report for simple or systematic with post-stratification with the thematic map
- Fix random seed value when it is automatic
- Fixbug restoring the response design window: while saving, first check if the response_design_window is opened, if so save some of the response_design_window configuration. Bug reported by Jhonatan Arias
- Fixes several variable names in the stratified and post-stratification items
- Fix the way to compare the thematic and post-stratication map
- Fixed active/inactive boxes depending on which Acatama dialogs are open and when they are closed
- Fixes typos in post-stratification vars, functs and texts.
- Fixes long and nan nodata values, and show full value of the integer and not in scientific format
- Fixes and getting the format of the nodata values for processing
- Fix number of samples based on minimum samples per stratum when the table is filling
- Fixes typos in post stratification items for the yml Acatama status file
- Fixed a bug when a pixel is being validated outside the post-stratification map
- Fixed a division by zero bug and guaranteed a minimum zoom value of 0.1
- Fix some weird RuntimeError when rubber band has been deleted from old instances of the plugin
- Updated the documentation
24.12:
- New dialog tool for estimating the appropriate sample size for simple random sampling using the equation by Stehman and Foody (2019), and Cochran (1977)
- New dialog tool for estimating the appropriate sample size for systematic random sampling, considering the valid map data and map using the equation by Stehman and Foody (2019), and Cochran (1977)
- Move the grid settings into a dialog to set the number of views for the response design window
- Make the main dock widget scrollable to improve usability on small screens
- Fixed auto fill the labeling buttoms from layer function, added another way to compute the unique values if no xml color style is defined in the layer
- Don't delete the thematic map instance layer when closing the sampling design window
- Fix an issue when the thematic map has not nodata value or is not set while computing the point spacing
- Improving the process when the tests are restoring the configuration file
- Added tests for number of samples and point spacing when determining the sample size
- Fix testing, do not update progress bar if pytest is running
- Save and restore the configuration for the new dialog tool for estimating the appropriate sample size for simple and systematic random sampling
- Fix a bug in the sampling report for simple and systematic sampling without post-stratification
- Rename a variable to avoid confusing it with another variable with a similar name
- Adjusts some dialog sizes
- Fix several warning and info boxes
- Set a warning message of the Qgis version needed for the latest Acatama version
- Improving the Acatama description
- Catch any issue while converting the path
- Check the path before compute the relative path
- Update the minimum Qgis version required by the plugin, issue #22
- All commits in: https://github.com/SMByC/AcATaMa/commits

tags=accuracy,accuracy assessment,thematic,classification,sample,sampling design,points,remote sensing
Expand Down

0 comments on commit c641f7a

Please sign in to comment.