Mask creation from quality pixel classification image

I have done the atmospheric correction in Sentinel-2 tile leading to atmospherically corrected S2A tile using Sent2Cor. Now, as classified by quality scene classification obtained from atmospherically corrected tile, I want to mask everything except vegetation in that image. Do we have any way to make a mask out of that quality classification scene?
Further, when I open the corrected product in SNAP I can find Quality scene classification image but when I go to the folder of corrected tile to find the same image(quality scene) so that I can use it to make a mask in ENVI, I couldn’t.
Does anyone have a way to solve this problem?
Thank you in advance.

you can use the valid pixel expression of the rasters (found in each band’s properties)

quality_scene_classification == 4

Thank you, Andreas. Now I want to assign no values to all the remaining entities except vegetation so that I can get only vegetations. Before what I did was, I assigned vegetation as 1 and other as 0 and applied it as a mask but now because of that 0 value, all the calculations are messed up. Could you tell me how to deal with this?

you can do all of this in one step in the band maths:

IF quality_scene_classification == 4 THEN 1 ELSE NaN

I did that but when I opened the image in arc Gis after that , the image is still 0 and 1.

that NaN definition only works for SNAP because it is thefined there what value stands for NaN
Every pixel must have a given value, either 0 or -9999 or -32768… It is defined in each software how these are interpreted, it cannot be ‘nothing’.

In ArcMap you also can determine the transparent pixel value in the layer properties under symbology.
Alternatively, you can perform a reclass and assing “NoData” to 0 which basically does the same as the formula above but for ArcMap.

Thank you. Its working.
Thank you very much.