Version 2 of FuzzyAttributes extends the plugin’s capabilities by directly integrating fuzzy transformation of rasters, in addition to vector attributes and text → fuzzy mapping. This allows you to work on continuous data (e.g., indices, altitudes, precipitation) and convert each pixel into a degree of membership between 0 and 1 according to a chosen fuzzy function.
Raw values are like the ingredients in a recipe: each has its own units, intensities, and tastes. Fuzzy values, on the other hand, reflect how we feel about these ingredients—what we like or dislike depending on the context. Working with degrees of satisfaction therefore means moving from a cold measurement to a nuanced assessment that is better suited to real-world decisions.
Raw values describe a physical phenomenon, but not its interpretive value for decision-making.
The transition to a satisfaction scale (0–1) allows the subjectivity of the decision-maker or expert to be introduced: it is no longer “what is the value?”, but “to what extent is this value appropriate?”.
This approach then makes it possible to coherently combine heterogeneous criteria (distance, altitude, precipitation, etc.) in fuzzy multi-criteria analyses.
Prepare the raster
- The plugin works on single-band rasters.
- Make sure your raster is loaded in QGIS.
- Check its CRS and extent, as these settings will be retained in the fuzzy raster created.
Launch FuzzyAttributes
- Menu: FuzzyAttributes > FuzzyRaster
- Select the source raster from the drop-down list.
- Choose the fuzzy function that suits your data:
- Linear increasing/decreasing
- Triangular
- Trapezoidal
- Sigmoid S or Z
- Gaussian
- Set the thresholds or limits for your fuzzy function.
- View the example graph in the dialog box to check that the transformation meets your expectations.
Create the blurred raster
- Click OK.
- By default, the output raster is named fzy_<raster_name>.tif.
- If a file with the same name exists, the plugin asks if you want to overwrite it or create a new version with an index.
The resulting raster retains the dimensions, CRS, and extent of the original, but each pixel is now a fuzzy value between 0 and 1, ready to be used for fuzzy analysis or spatial aggregation.
Viewing the result
- The resulting raster is automatically loaded into the current project.
- By default, a graduated symbology is applied to see the variations in fuzzy values.
- Combine it with other rasters or vector layers to continue your fuzzy analyses.
Next steps with FuzzyAttributes V2
The transformation of a raster into fuzzy values is the first module in version 2.
Other new features include:
- Raster Aggregation: combine two fuzzy rasters according to a subjective function.
- Classes → Fuzzy: reclassify a raster of discrete classes into fuzzy values using a correspondence table.
- Multi-criteria aggregation and analysis with a graphical interface and consistency checking.
Tip: You can chain fuzzy conversion and aggregation. pour créer des analyses multi-critères directement dans QGIS, sans exporter vos données vers d’autres logiciels.
