Chopping whilst scanning results in an image that contains two beams (a plus and minus image of the source). To restore the source profile we must deconvolve the chop from the measured map. The problems associated with this step can best be appreciated by considering the Fourier transform (FT) of the chop function, which is a sine wave with zeroes at the origin and at harmonics of the inverse chop throw. Deconvolving the chop function is equivalent to dividing the FT of the measured map by the FT of the chop and then transforming back to image space. Clearly, problems arise at spatial frequencies where the sine wave of the chop FT has a low value or is zero. Noise at these frequencies is blown up and significantly reduces the signal-to-noise of the restored map [11].
SURF -- SCUBA User Reduction Facility