Input file names (can be byte, halfword, I*4, or R*4)
Output file names (can be byte, halfword, I*4, or R*4)
The standard VICAR size field: starting line, starting sample, number of lines, and number of samples (if MSS is specified, the size field refers to one band, not to the entire image.)
MATRIX = (M11,M12...) The matrix values are real numbers and are entered in row order. The column number is the number of input data sets or bands, the row number is the number of output data sets or bands
Valid keyword values: BYTE, HALF, FULL, REAL. This specifies the output data format.
MSS=NI Denotes that the input data set is in MSS format with NI interleaved inputs. limit: NI <= 32
If the input is a single file of multichannel data, the USEBANDS parameter is used to specify which bands are to be used as input, and the order that must be used to be consistent with the MATRIX parameter. The default is to use all bands, in ascending order.
Offset = 0 and gain = 1 for all output bands.
GAIN=Q For all output bands,
or GAIN=(Q1,Q2,Q3,...) The nth output band is scaled by gain Qn.
OFFSET=P For all output bands,
or OFFSET=(P1,P2,P3,...) The nth output band is scaled by offset Pn.
AREA=(SL,SS,NL,NS) Indicates area to be sampled for auto-scaling.
INC=N For auto-scaling, every Nth sample of every Nth line is used.
LINC=N For auto-scaling, every Nth line is used.
SINC=N For auto-scaling, every Nth sample is used.
PERC=P P percent saturation is to be allowed in output histograms, one-half at each end. (default=1)
LPER=L L percent saturation is to be allowed in output histograms at the low end. (default=0.5)
HPER=H H percent saturation is to be allowed in output histograms at the high end. (default=0.5)
SPREAD=S Output histograms are to have a spread of s. (default=full range of data type) See also parameter FIRM.
Specify FIRM to cause the output to be clipped so that SPREAD and CENTER apply firmly. Otherwise (the default) SPREAD and CENTER apply approximately.
CENTER=C Output histogram is to be centered about c. (default=SPREAD/2 for byte data, =0 otherwise.)
EXCLUDE=E This keyword instructs the program to ignore input pixels of value e when determining saturation points. The algorithm used simply zeros the histogram bin which results from each input DN level e. However, other combinations of input DNs will in general also contribute to this bin and will be erroneously ignored. Therefore, this keyword should only be used if a large contribution from spurious value (e.g., zero) is expected.
In the auto-scale mode, a histogram is first formed to determine the range of the output data without rescaling. This parameter allows the user to specify the number of bins used to form the histogram of each output band. While this feature is rarely needed for byte data, the user may want to adjust this (and the related RANGE parameter) for other data types, if the unscaled output range is uncertain.
In the auto-scale mode, a histogram is first formed to determine the range of the output data without rescaling. This parameter allows the user to specify the range of the bins used to form the histogram of each output band. While this feature is rarely needed for byte data, the user may want to adjust this (and the related HISTSIZE parameter) for other data types. The auto-scaling algorithm works best when the histogram bins span only the actual range of the data. Therefore, if it is known that the transformation will produce values only within a certain range, that range should be specified here.