A simple matrix method and computer program product for stray-light correction in imaging instruments is provided. The stray-light correction method includes receiving raw signals from an imaging instrument and characterizing the imaging instrument for a set of point spread functions. For high resolution imaging instruments, the raw signals may be compressed to reduce the size of the correction matrix. Based on stray-light distribution functions derived from the point spread functions, a correction matrix is derived. This fast correction is performed by a matrix multiplication to the measured raw signals, and may reduce stray-light errors by more than one order of magnitude. Using the stray-light corrected instrument, significant reductions may be made in overall measurement uncertainties in radiometry, colorimetry, photometry and other applications. Because the PSFs may include other types of undesired responses, the stray-light correction also eliminates other types of errors, e.g., interreflection between a CCD and the detector window.