A system for the photometric calibration of streak cameras and similar imaging devices provides a precise knowledge of the camera's flat-field response as well as a mapping of the geometric distortions. The system provides the flat-field response, representing the spatial variations in the sensitivity of the recorded output, with a signal-to-noise ratio (SNR) greater than can be achieved in a single submicrosecond streak record. The measurement of the flat-field response is carried out by illuminating the input slit of the streak camera with a signal that is uniform in space and constant in time. This signal is generated by passing a continuous wave source through an optical homogenizer made up of a light pipe or pipes in which the illumination typically makes several bounces before exiting as a spatially uniform source field. The rectangular cross-section of the homogenizer is matched to the usable photocathode area of the streak tube. The flat-field data set is obtained by using a slow streak ramp that may have a period from one millisecond (ms) to ten seconds (s), but may be nominally one second in duration. The system also provides a mapping of the geometric distortions, by spatially and temporarily modulating the output of the homogenizer and obtaining a data set using the slow streak ramps. All data sets are acquired using a CCD camera and stored on a computer, which is used to calculate all relevant corrections to the signal data sets. The signal and flat-field data sets are both corrected for geometric distortions prior to applying the flat-field correction. Absolute photometric calibration is obtained by measuring the output fluence of the homogenizer with a "standard-traceable" meter and relating that to the CCD pixel values for a self-corrected flat-field data set.