In mixed mode HiRISE images, there is a tendency for every other line in the ccd with the lower binning mode to be significantly darker or lighter than normal image data would suggest. This program attempts to correct that problem. 2 passes through a cube are necessary to perform the correction. The first pass gathers information on each line of the cube to determine if the striping phenomenon occurs in odd numbered lines or in even numbered lines, and also gathers information to perform the actual correction.

The second pass through the data performs the actual correction, of which there are two different types. The default, an additive correction, is of the following form for each pixel

    					     
                                         ((Line[i-1]) + Line[i+1])
                Pixel[i,j] - Line[i] +   --------------------------
                                                      2
    

The other correction, a multiplicative correction of a similar style, is of the form

                Pixel        ((Line[i-1]) + Line[i+1])
               -------   *  ---------------------------
               Line[i]                    2
    

It should finally be noted that the striping pattern has been observed to be most prevalent in a certain quardrant of the image, though that phase is not the same for all images. In order to address this irregularitiy, the line averages shown above are not for the entire line, but only for the quarter of the line in which sample i exists.

Caution should be taken when running hidestripe: it should only be run on images where different CCDs are run at different binning modes. Of those, only channels that are not used in the highest binning mode of the image should be corrected. For example, if an image has binning modes 1, 2, and 4 the CCDs with modes 1 and 2 may be hidestripe'd, but in an image that is only binned in modes 1 and 2, the bin 1 CCDs are the only ones that need the hidestripe correction.

This example shows the default additive correction applied to a binned 0 image in an image that is mixed and includes binning modes 0 and 1