general.cpp

来自「基于块方向的图像无损压缩代码」· C++ 代码 · 共 113 行

/*++*
* Copyright (c) 1997 University of British Columbia. All rights reserved.
*
*       File:           general.c
*       Pourpose:       contains the general encoding routins (eg. Golomb encode).
*
*       Author:         Ismaeil R. Ismaeil, Aug. 1997
*
*--*/

#include "stdafx.h"
#include "globals.h"
#include "prototypes.h"

/********************************************************************
 Error quantization for near-lossless mode, and reconstructed value 
********************************************************************/
int Quantize(int buff)
{//针对有损而言
	if(buff > 0)
		buff = (buff+NEAR_LOSSLESS)/(2*NEAR_LOSSLESS+1);
	else
		buff = - (NEAR_LOSSLESS-buff)/(2*NEAR_LOSSLESS+1);
	
	return buff;
}

/**************************************
Operation for the near lossless mode.
**************************************/
int ComputeRx()
{
	Rx = Px + SIGN*Errval*(2*NEAR_LOSSLESS+1);
	if(Rx < 0)
		Rx = 0;
	else if(Rx>MAXVAL)
		Rx = MAXVAL;
	return Rx;
}

/*************************************
Modulo reduction of the error
*************************************/
int ModRange(int buff, int range)
{
	if(buff < 0)
		buff = buff + range;
	if(buff >= ((range+1)/2))
		buff = buff - range;
	return buff;
}

/************************************
Gets the next sample to be encoded 
************************************/
void GetNextSample()
{
	count++;
	if(x.y == dim.y)
	{
		x.x++;
		x.y = 1;
	}
	else
		x.y++;
 
	a.x = x.x;
	a.y = c.y = x.y-1;
	b.x = c.x = d.x = x.x-1;
	b.y = x.y;
	d.y = x.y+1;

	Ix = in_buf[x.x][x.y];
	Ra = in_buf[a.x][a.y];
	Rb = in_buf[b.x][b.y];
	Rc = in_buf[c.x][c.y];
	Rd = in_buf[d.x][d.y];

	if(x.y == dim.y)
		EOLine = 1;
	else
	EOLine = 0;
}

/***************************************
Golomb encode the maped error values  
***************************************/
void GolombCoding(int n, int l, int limit)
{
	int val1;
	int val2;
	int val3;
	int val4;

	val1 = n >> l;
	val2 = val1 << l;
	val3 = limit-qbpp-1;
 
	if(val1<val3)
	{
		AppendToBitStream(0,val1);
		AppendToBitStream(1,1);
		val4 = n - val2;
		AppendToBitStream(val4,l);
	}
	else
	{
		AppendToBitStream(0,val3);
		AppendToBitStream(1,1);
		AppendToBitStream(n-1,qbpp);
	} 
}