You might want to use the QuickBird Radiance tool under Basic Tools ->Preprocessing->Calibration Utilities. This tool will convert the scaled integers or byte values to floating point radiance in units of um/(cm^2*sr*um).

Thanks for the reply. I already perform this step when I'm converting to top-of-atmosphere radiance. I get values that range from 0 to 1169.00 uW/(cm^2*nm*sr), which are in floating point values. I believe typical values might range from around 0 to 16 uW/(cm^2*nm*sr). I'm just wondering if I would have to any re-scaling from 16 bit to 11 bit (DN 0-2048) to get the appropriate range of values of uW/(cm^2*nm*sr). Thanks

Where did you get your data? While 11-bit QuickBird data is stored as 16-bit integers, the dynamic range of the data is still typically 0-2047. I have not seen data distributed by DigitalGlobe that uses values up to 65536. It makes me think some other processing has occured on your dataset to actually use that range.. If you are in units of uW/(cm^2*nm*sr) then the values would range from around 0 - 35 at most. I would confirm the range of data with your data provider and ask if there was some other sort of scaling or processing that was applied.

Thanks for the reply. The images I have obtained were othrorectified, so I did the processing on the original .ntf files and the output looks good. Your right, some kind of processing has occurred that I'm not aware of. I will need to contact the image provider.