// = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
//      THREEE DIMENSIONAL SPIHT COMPRESSION FOR IMAGE SEQUENCES
//            
// = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
//
// Copyright (c) 1995, 1996, 1997 by  Beong-Jo Kim, William A. Pearlman,
// and Amir Said.

// These programs are protected by Copyright (c) and patents.
// They may not be redistributed without the consent of the copyright
// holders. In no circumstances may the copyright notice be removed.
// The programs may not be sold for profit nor may they be incorporated
// in commercial programs without obtaining a license.
// For licensing information, please contact
//  Dr. William Pearlman, pearlman@ecse.rpi.edu, (518) 276-6082.

// These programs are provided as is, without any express or
// implied warranty, without even the warranty of fitness for a
// particular purpose.

// - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

// Note:  The routines (algorithms) may have bugs, despite our best efforts 
	to remove as many as possible. Also, the programs have 
	never been optimized to minimize computational time.
//-----------------------------------------------------------------------------------
This program compresses both 1 and 2 bytes/pel/frame raw gray scale image sequences in 16 frame segments by the 3D SPIHT algorithm.	

It has been tested on a Linux machine and should work similarly for a Windows/DOS machine. 

This program assumes the input file is raw file format.
This program is working with 1 byte per pixel and 2 byte per pixel grayscale images. 
Input sequence should be raw file format
   with frame naming convention sequence_name001, sequence_name002,
   ..., etc.  There should no .raw extensions.  The keyword for the 
   sequence is sequence_name. 

For 2 byte per pixel images, the first byte should be the higher 8 bits, 
  and the second byte should be the lower 8 bits.

Usage :
For usage info, type "codetree3D -u"

Encoding :
codetree3D keyword coded_file first_f last_f pel_bytes(1/2) bpp_rate [smoothing] width height
   : keyword - keyword of image file name.
   : coded_file - compressed file name
   : first_f - first frame number
   : last_f - last frame number
   : pel_byte - number of byte per pixel (1 or 2)
   : bpp_rate - bit per pixel (bpp), type 0 for lossless compression
   : width - number of pixels per line
   : height - number of lines per pixels

Decoding :
decdtree3D coded_file original_sequence_keyword first_f last_f bpp_rate 
   : compare the decoded sequence with original sequence	   
	   
decdtree3D -s  coded_file new_sequence_keyword first_f last_f bpp_rate 
   : use -s command line option to produce decoded sequence files
 
Example: Frames 0 to 15 of the sequence "football", rate of 1.0 bits/pel.       

Encoding:
codetree3D football football.cmp 0 15 1 1.0 512 512
  : the program assumes the image file names would be
    football000
    football001
    football002
        :
        :
    football015 (all raw format)
    ,with compressed file name of "football.cmp"

Decoding:
decdtree3D [-s] football.cmp football 0 15 1.0
where football is reconstructed image sequence and football.cmp is compressed image sequence.