.TH QCCHYPRKLTTRAIN 3 "QCCPACK" "".SH NAMEQccHYPrkltTrain,QccHYPrkltFactorization,QccHYPrkltTransform,QccHYPrkltInverseTransform\- training, factorization, and forward/inverse transform forthe reversible KLT.SH SYNOPSIS.B #include "libQccPack.h".sp.BI "int QccHYPrkltTrain(const QccVolumeInt " image ", int " num_bands ", int " num_rows ", int " num_cols ", QccHYPrklt *" rklt );.br.BI "int QccHYPrkltFactorization(QccHYPrklt *" rklt );.br.BI "int QccHYPrkltTransform(QccVolumeInt " image ", int " num_bands ", int " num_rows ", int " num_cols ", const QccHYPrklt *" rklt " );.br.BI "int QccHYPrkltInverseTransform(QccVolumeInt " image ", int " num_bands ", int " num_rows ", int " num_cols ", const QccHYPrklt *" rklt );.SH DESCRIPTION.BR QccHYPrkltTrain()determines the KLT transform mean vector and transform matrixfor the pixel vectors contained in.IR image .Together, the mean vector and transform matrix constitute the KLTfor.IR image ,and they are returned in.IR rklt ..IR rkltmust be allocated prior to calling.BR QccHYPrkltTrain() ;furthermore, the.IR num_bandsfield of.IR kltmust match the.IR num_framesfield of.IR image(i.e., the number of frames of the image cube is considered to bethe number of spectral bands of the hyperspectral image).The operation of.BR QccHYPrkltTrain()is essentially the same as that of.BR QccHYPkltTrain (3);the primary difference is that the mean vectoris integer-valued as is needed for a reversible KLT..LP.BR QccHYPrkltFactorization()factors the KLT matrix into the lifting-step matrices. See.BR QccHYPrklt (3)for more details on this process..BR QccHYPrkltTrain()must be called prior to calling.BR QccHYPrkltFactorization() ..BR QccHYPrkltFactorization()sets.IR rklt -> factored ..LP.BR QccHYPrkltTransform()applies the forward reversible KLT spectrally to each pixel vector in.IR image ,which is overwritten with the transformed dataset..BR QccHYPrkltFactorization()must be called on.IR rkltprior to calling.BR QccHYPrkltTransform()which checks to see if.IR rklt -> factoredis set and returns in error if it is not..LP.BR QccHYPrkltInverseTransform()is the corresponding inverse reversible transform..BR QccHYPrkltFactorization()must be called on.IR rkltprior to calling.BR QccHYPrkltInverseTransform()which checks to see if.IR rklt -> factoredis set and returns in error if it is not..SH "THEORY"See.BR QccHYPrklt (3)..SH "RETURN VALUE"Both routinesreturn 0 on success, 1 on error..SH "SEE ALSO".BR QccHYPrklt (3),.BR QccHYPkltTrain (3),.BR QccPackHYP (3),.BR QccPack (3).SH AUTHORCopyright (C) 1997-2009  James E. Fowler.\"  The programs herein are free software; you can redistribute them an.or.\"  modify them under the terms of the GNU General Public License.\"  as published by the Free Software Foundation; either version 2.\"  of the License, or (at your option) any later version..\"  .\"  These programs are distributed in the hope that they will be useful,.\"  but WITHOUT ANY WARRANTY; without even the implied warranty of.\"  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the.\"  GNU General Public License for more details..\"  .\"  You should have received a copy of the GNU General Public License.\"  along with these programs; if not, write to the Free Software.\"  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.