diff options
Diffstat (limited to 'noao/imred/vtel/doc/unwrap.hlp')
| -rw-r--r-- | noao/imred/vtel/doc/unwrap.hlp | 95 |
1 files changed, 95 insertions, 0 deletions
diff --git a/noao/imred/vtel/doc/unwrap.hlp b/noao/imred/vtel/doc/unwrap.hlp new file mode 100644 index 00000000..67fad069 --- /dev/null +++ b/noao/imred/vtel/doc/unwrap.hlp @@ -0,0 +1,95 @@ +.help unwrap May87 noao.imred.vtel +.ih +NAME +unwrap -- Filter an IRAF image; remove binary wrap-around. +.ih +USAGE +unwrap listin listout +.ih +PARAMETERS +.ls listin +List of images to unwrap, this is an IRAF template. +.le +.ls listout +List of output images, this is an IRAF template. If the output list +is the same as the input list, the unwrapping is done in-place. +.le +.ls threshold1 = 128 +Data jump threshold for first unwrap pass. +.le +.ls wrapval1 = 256 +Factor to multiply wrap value by for first unwrap pass. +.le +.ls threshold2 = 128 +Data jump threshold for second unwrap pass. +.le +.ls wrapval2 = 256 +Factor to multiply wrap value by for second unwrap pass. +.le +.ls cstart = 2 +Column of image to start unwrapping. Columns are numbered from left to right. +.le +.ls step = 5 +Number of steps (1-5) to perform on image (unwrap1, difference, unwrap2, +reconstruct, fixlines). +.le +.ls verbose = yes +If set, program produces progress reports, etc. +.le +.ih +DESCRIPTION +Unwrap checks for binary wraparound in IRAF images. +The algorithm consists of reading the image line by line, unwrapping +each line, and writing the line out to another image. The procedure +for unwraping is a five step process. +.ls Step one: unwrap1 +Unwrapping is accomplished by scanning the data line and looking for +large jumps in the data values. Large negative jumps are interpreted +as data wrapping and large positive jumps are interpreted as data unwrapping. +The program keeps track of the number of wraps, each data element in the +array has wrapval1 * wrapnumber added. This effectively unwraps an image +in which the point to point variation in the data values is small compared +to the variation caused by a binary wrap. +.le +.ls Step two: difference +A difference image is produced from the above step one image by calculating +the pixel to pixel difference between all of the pixels in the line. The +first column of the image is generally left intact so that the image can +be reconstructed in a later step. Step one often produces streaks in the +image due to data variation large enough to mimic wrapping. This step +two difference image eliminates most of these streaks except for their +point of origin, where the confusion occured. +.le +.ls Step three: unwrap2 +This is the second unwrapping step. The image is unwrapped as in step +one using the second set of unwrap values (threshold2, wrapval2). +.le +.ls Step four: reconstruct +The original image is reconstructed from the step three image by +adding pixel values successively to line pixels. +.le +.ls Step five: fixlines +If bad lines (streaks) still can be found in the image, they are +eliminated by replacing the line by the average of the lines above +and below bad line. +.le +.ih +EXAMPLES +1. To unwrap an image called "continuum" and store the resulting image in +"unwrapped", and use the default parameters, the command might be: + +.nf + vt> unwrap continuum unwrapped +.fi + +2. To unwrap all the images in the directory starting with s1492 and store +the unwrapped images in s1492*u, to start in column 31, to do four steps, +and to see verbose output, the command might be: + +.nf + vt> unwrap s1494* s1492*//u cstart=31 step=4 v+ +.fi + +.ih +SEE ALSO +.endhelp |