Transform image from SNAP to QGIS

Dear all,

I have a question that maybe it is more technical.

I produced an image in SNAP by using (phase to elevation- InSAR method). I plan to label it to use in a manuscript but I have some questions about it.

SNAP automatically lables image. In my case; it is from 5.59 to 8.84 (figure 1) but when I use (Analysis……statistics); then It shows me a min and max values that are from 2.64 to 12.67(figure 2)



Figure 2

Now I do not know what should I use for labling?

Another thing;

When I transform image from SNAP to QGIS, then QGIS automaticly gives it min and max (for example here is from 0 to 12.67 (figure 3));

So what is the correct one and should be used in publication?

Another question;

I want to use stepwise breaks in my colorbar, not 4.98 etc. but when min is ……and max is ……then how can put another number there?


the min/max values from the statistics menu are correct. The color lables are sometimes clipped off (2.5%) to enhance the contrasts.

QGIS does the same: Extreme values are not used in the coloring. You can change that in the “Min/Max Value Settings”

Thank you so much @ABraun for reply but then I do not know why in QGIS is from 0 to 12.67 but in SNAP (histogram) is from 2.64 to 12.67?

and another thing;
I want to use stepwise breaks in my colorbar, not 12.67 for maximum but when max is 12.67 then how can put another number there?

Did you export the elevation to another file format? DIM to TIF maybe?

Change to Equal Interval mode, then adjust the number of classes, then change the values in the first column according to your needs.

yes, I used dim to tiff.

I know how change it but the problem is; if my real maximum is 12.67 then is it OK to use 12 in labling? because 12 is not real maximum.

this possibly changed your raster values, at least this is reported a couple of times in here.
I suggest to directly open the img file in QGIS without converting to GeoTiff

The generalization of the legend is heavily discussed. I prefer a more readable legend but some scientists insist on precise labels instead. Both has pros and cons.

Thank you so much @ABraun . Yes my co-author prefer this tepwise :smiley:. Thanks helpful man.

Dear @ABraun

As you mentined before:

the min/max values from the statistics menu in SNAP are correct. The color lables are sometimes clipped off (2.5%) to enhance the contrasts.

But when you look at same image; one time with min and max values and another time with clipped off (2.5%); it looks with clipped off (2.5%), image looks better to visual and to eye.

Is it possible to use it in our publications instead of min and max values?


For visualization, I would say yes (as long as you mention it).
For data analysis I don’t think it is a good idea.

1 Like

Dear @ABraun
Sorry. One more question regarding this:
please have a look in below:

As you can see min and max values are between (-10.98) to (7.07) but based on pixel-value plot; most of pixels are between almost (-1) to (1). So what is your suggestion to use in visualization?

I’m sorry, I cannot give a recommendation here based on this alone.
Often 2 standard deviations are used (sigma=0.520) for the color range, so +/- 1.06 is probably a good start. But you have to play around with the values and test how the contrasts are and how well the image supports what you want to demonstrate with it.

It is probably a better idea to talk about this with your supervisor.

1 Like