NORMALISATION allows non-uniform sensitivity response to be corrected for detector systems which suffer from non-uniform response in a manner which is equal (or at least similar) for all rows or all columns.
The method is to mask-off all data except for the background data which are (ideally) assumed to be approximately flat. The program adds up all the values in each column or row, and produces a 1-D normalisation array. This array is then used to correct all the data in each column or row. The output is in the memory. The user can select the direction of the correction.
(Clearly this is only an approximate correction, and if available a calibration image should be used to apply the flat-field correction.)