I think you would need to use masking but you can only build a mask based on one band in a file. There is an option when building a mask based on the Data Range to 'mask pixels if ANY band matches range'. This would create a single mask of pixels from ANY band that meet the min/max criteria and might give you what you need. Otherwise, I think you would have to create a mask or ROI based on thresholds or data range for each band.

My problem is that the pixels that i want to be zero have the value of 1.5 of reflectance. As a result when i' m trying to apply mask or another technique it gives me wrong results. First i want to replace those values with the zero at those specific pixels and then i' ll try to apply or build a mask!Any advice for that?Thank you very much for your answer. It was very helpful!

Yes, use Build Mask to mask pixels with a max value of 1.5. Then apply the mask. When you apply the mask, you can choose the value of the masked pixels (0) or have them be transparent by setting them to the Data value to ignore. You can also create an ROI based on thresholds to make sure you are finding the right pixels, then use the ROI in Build mask.

OK!Thanks for your reply!I found another way with the band math. I wrote at the band math (b1 gt 1)*0+(b1 lt 0)*0+(b1 ge 0 and b1 le 1)*float(b1). I think that it works perfect!