lincont Documentation

lincont - Linear continuum removal of each spectrum
"lincont" performs a linear continuum removal on each spectrum.  The
slope for each spectrum is computed and applied to linearize the
spectrum. "lincont" allows the user to select which method to apply to
compute the continuum - SUBTRACTION, RATIO or BANDEPTH.  The basic
difference between the SUBTRACTION method and the RATIO method is that
the continuum of each input DN is computed by subtracting the incoming
DN by the linearization parameter or dividing it by the parameter,
respectively.  The BANDEPTH method is a classical band depth
computation where the reflectance of the band is subtracted from the
reflectance of the continuum line and the result is divided by the
reflectance of the continuum line.


The user must specify two bands numbers from the input cube that must be
selected by the input subcube specifier that will be used as two points
to compute the slope.  Note that these band numbers correspond to the
band ordering index of the input cube and not the original band numbers.
These two points in each spectrum are used to compute the equation

y = mx + b, where
m = (DN2 - DN1)/(Wavelength2 - Wavelength1), and
b = DN1 - m * Wavelength1.

Each new DN is computed as
DNout(i,j,k) =   DNin(i,j,k) - Y(i,j,k) + ADDB, (SUBTRACTION) or
DNout(i,j,k) = { DNin(i,j,k) / Y(i,j,k) + ADDB, (RATIO)
               { NULL if Y(i,j,k) = 0.0
DNout(i,j,k) = { (Y(i,j,k)-DNin(i,j,k))/Y(i,j,k) + ADDB, (BANDEPTH)
               { NULL if Y(i,j,k) = 0.0

Y(i,j,k)     = m(i,j,k) * Wavelength(i,j,k) + b(i,j,k)
m(i,j,k)     = (DN(i,j,Wavelenght2) - DN(i,j,Wavelenght1)) /
               (Wavelength2 - Wavelength1)
b(i,j,k)     = DN(i,j,Wavelength1) - m(i,j,Wavelength1) * Wavelength1

Note that if either of the 2 DN values are special pixel values, the
slope is indetermined and the entire spectrum is NULLed.  The number of
spectra reported NULLed at the end of the program are only those spectra
that contained valid pixels, not spectra entirely composed of special
pixel values (these spectra are set to NULL, however).  Also note it is
important to ensure the values in the BAND_BIN_CENTER keyword in the
BAND_BIN group are accurate.

Note that the user may OPTIONALLY supply his own wavelength values via
the WAVLEN parameter.  If the values in the BAND_BIN_CENTER keyword in
the label are sufficient, the user should set these values to null (the
default).

If the wavelengths from the label are used, then the DN values of both
bands used to compute the slope will attain the value of ADDB.  All
other spectra will be centered (normalized) about the value of ADDB.

Programmer:  Kris Becker, U.S.G.S., Flagstaff, Az

Input cube file name
  (default extension is .CUB)

NONE

Input subcube specifier

" "

Output cube file name
  (default extension is .CUB)

NONE

Output pixel type

0

Output pixel data range

0.0,0.0

The bands used to compute the
slope for each spectrum

NONE

Optional wavelengths to use for
each band specified in LBANDS

--

Method used to compute the
continuum
SUBTRACTION
RATIO
BANDEPTH

"BANDEPTH"

Additive constant applied to
each spectrum after
linearization

0.0

Comment for user

" "

ADDITIONAL NOTES:

Parm	Description
FROM	The input cube file name that is to be normalized. EX: ISIS$WORK:fromfile.CUB
SFROM	Input subcube specifier. Selects the part of the input cube file that is to be normalized.
TO	Name of the output normalized cube file. The dimensions of this file are subject to the input subcube specifier applied to the FROM file. EX: ISIS$WORK:tofile.CUB
OTYPE	Output pixel data type. Permitted values are: 0 - output type is same as input file pixel type 1 - 8-bit (integer with type conversion parameters) 2 - 16-bit (integer with type conversion parameters) 3 - 32-bit (floating point)
ORANGE	Output pixel data range. If the output pixel type is 1 (8-bit integer with type conversion parameters) or 2 (16-bit integer with type conversion parameters), then the type conversion parameters in the output file will be set to values that allow representing the specified range of output values. Output values outside this range will be stored as the special "representation saturation" value. The ORANGE parameter is ignored if the output pixel type is 3 (32-bit floating point) since type conversion parameters are not applicable to floating point pixel values. If both ORANGE(1) and ORANGE(2) are 0.0, then the type conversion parameters in the output file will automatically be set to allow representing the same range of values as can be represented in the input file. (The user will be required to supply a specific range for ORANGE if the input pixel type is 3 (32-bit floating point) and the output pixel type is 1 (8-bit with type conversion parameters) or 2 (16-bit with type conversion parameters)).
LBANDS	LBANDS selectes the two bands the which to compute the slope with. Note that the coresponding wavelenghts and DN values in each spectrum will be used to compute the slope and intersect in each spectrum.
WAVLEN	This parameter allows the user to optionally select new wavelengths for each band selected in LBANDS, respectively. Note that if this option is used, these values are used to overide the wavelengths values found in the label in the BAND_BIN_CENTER keyword for calculation of the wavelength contribution of the slope ONLY.
METHOD	METHOD allows the user to select the algorithm used to compute the linear continuum. SUBTRACTION will use a difference algorithm to compute the continuum whereas RATIO will use a division algorithm. BANDEPTH applies a classical band depth computation to generate the result. See the general help for more details.
ADDB	An additive constant supplied to the user that is applied to each band in the spectrum after the linearization is performed. This allows the user to center each spectrum at this value.
USERNOTE	Specifies a comment that will be entered into the output log file and history entry in the output file.