Sharpening the Image

To sharpen, first create a mask to exclude the sea from sharpening. This broadly corresponds to the pixels with medium brightness. Click the top layer, and then choose Select | Select from Color Range. In the dialog box, click Shadow in the drop-down list at the top left, and then click Select and OK.

Now call the Select from Color Range function again, but create a selection for the highlights this time. Make sure that the Add to Current Selection option is selected. Then click on the new white layer in the Layers panel. The unselected areas will appear in red in the preview window. This gives you a better approach to adjusting the mask's blur.

Then select the Gaussian Blur filter and choose a value of about 50 pixels for the radius. After pressing OK to confirm, click on the layer below the mask again. Create a new filter mask layer for it and select Sharp from the Improve group as the filter.

Last but not least, save the final edited image in two file formats: Krita's own image format, which contains all layers, and a common graphics format, such as JPEG or PNG. You can call the export dialog by selecting File | Export (Figure 14).

Figure 14: The final results prove that Krita is excellent for editing images.

The Krita file gives you the ability to easily change the editing steps without having to go through the entire editing process again. When saving to other graphics formats, make sure that Krita does not automatically convert the image to the sRGB color space.

Conclusions

Krita and the G'MIC filter collection are excellent for editing photos. However, this combination is more complicated to use and editing usually takes longer than with a RAW developer such as RawTherapee or darktable. Krita as a photo editor is especially useful for users who are not comfortable with darktable. In some cases, the retouching functions of RAW developers are too rudimentary. Krita is also recommended for experienced photographers who are looking for a free Photoshop alternative and prefer the layer-based old-school technique, but at the same time do not want to do without large color spaces and adjustment layers.