I have a raster GeoTIFF with values from 0 to 255 representing intensities. I want to Vectorize this information, but in ranges of 25. As an example, I want to create a layer in which polygon islands representing values grouped between 230-255; 25-50; etc.

In another thread someone recommended using the "Vectorising Grid Classes" but that only works for 1 value!


QGIS 2.18

If you are using QGIS 2.18, you can use Reclassify values under SAGA in Processing toolbox

  • Go to Processing toolbox → SAGA → Raster tools → Reclassify values

enter image description here

  • Select raster file that you want to classify
  • Method: Simple table

enter image description here

  • Lookup Table click on the a new window will open

enter image description here

  • In the example, the raster data were classified into six classes, so you need to change the values based desired range and output

  • Reclassified grid: write a name for the new classified raster data.

Please make sure that you have selected the correct projection when you created the slope percent to give you the desired output from reclassify tool.

QGIS 3.4

In QGIS 3.4.1, there is a Reclassify by table tool which is located under Processing toolbox -> Raster analysis -> Reclassify by table.

enter image description here

Using the tool is almost same as the one explained for SAGA Reclassify tool.

After reclassify the data, you can use Vectorize tool to convert the raster into vector data. The Vectorize tool exists in both QGIS 2.18 and 3.4.

  • I followed these instructions for QGIS 3 but I get a "bad allocation" error – Steffan Clent Davies Nov 30 '18 at 9:47

First reclassify raster into desired clasess then use vectorize to convert it into vector. After you might want to simplify vector to get smoother lines.

  • I get a "bad allocation" error after reclassification – Steffan Clent Davies Nov 30 '18 at 9:48
  • What version of QGis, and which tool did you use. You can try grass or saga reclassify tools in 2.18, QGIS3 has native tool.. – Mat Nov 30 '18 at 10:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.