1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
|
/*************************************************************************
This project implements a complete(!) JPEG (Recommendation ITU-T
T.81 | ISO/IEC 10918-1) codec, plus a library that can be used to
encode and decode JPEG streams.
It also implements ISO/IEC 18477 aka JPEG XT which is an extension
towards intermediate, high-dynamic-range lossy and lossless coding
of JPEG. In specific, it supports ISO/IEC 18477-3/-6/-7/-8 encoding.
Note that only Profiles C and D of ISO/IEC 18477-7 are supported
here. Check the JPEG XT reference software for a full implementation
of ISO/IEC 18477-7.
Copyright (C) 2012-2018 Thomas Richter, University of Stuttgart and
Accusoft. (C) 2019-2020 Thomas Richter, Fraunhofer IIS.
This program is available under two licenses, GPLv3 and the ITU
Software licence Annex A Option 2, RAND conditions.
For the full text of the GPU license option, see README.license.gpl.
For the full text of the ITU license option, see README.license.itu.
You may freely select between these two options.
For the GPL option, please note the following:
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it 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 this program. If not, see <http://www.gnu.org/licenses/>.
*************************************************************************/
/*
**
** Inverse DCT operation plus scaled quantization.
** This is an unscaled fix-point DCT. It requires approximately 45 shifts
** per row and column.
**
** $Id: idct.cpp,v 1.24 2016/10/28 13:58:54 thor Exp $
**
*/
/// Includes
#include "interface/types.hpp"
#include "std/string.hpp"
#include "dct/idct.hpp"
#include "tools/environment.hpp"
#include "tools/traits.hpp"
#include "interface/imagebitmap.hpp"
#include "colortrafo/colortrafo.hpp"
#include "marker/quantizationtable.hpp"
#include "coding/huffmancoder.hpp"
///
/// Defines
// Number of fractional bits.
#define TO_FIX(x) WORD((x * (1UL << FIX_BITS)) + 0.5)
// Actually, this is also simply a long, but prescaled by FIX_BITS
#define FIXED T
// This is also simply LONG, but with two bits prescaled
#define INTER T
// Backshift from FIX_BITS to INTERMEDIATE_BITS
#define FIXED_TO_INTERMEDIATE(x) (((x) + (1L << (FIX_BITS - INTERMEDIATE_BITS - 1))) >> (FIX_BITS - INTERMEDIATE_BITS))
// Preshifted by FIX_BITS + INTER_BITS
#define INTER_FIXED T
// Backshift from INTERMEDIATE+FIX bits to integer, plus 3 bits for the DCT scaling
#define INTER_FIXED_TO_INT(x) (((x) + (1L << (FIX_BITS + INTERMEDIATE_BITS + 3 - 1))) >> (FIX_BITS + INTERMEDIATE_BITS + 3))
///
/// IDCT::IDCT
template<int preshift,typename T,bool deadzone,bool optimize>
IDCT<preshift,T,deadzone,optimize>::IDCT(class Environ *env)
: DCT(env)
{
}
///
/// IDCT::~IDCT
template<int preshift,typename T,bool deadzone,bool optimize>
IDCT<preshift,T,deadzone,optimize>::~IDCT(void)
{
}
///
/// IDCT::DefineQuant
template<int preshift,typename T,bool deadzone,bool optimize>
void IDCT<preshift,T,deadzone,optimize>::DefineQuant(class QuantizationTable *table)
{
const UWORD *delta = table->DeltasOf();
int i;
// No scaling required here.
for(i = 0;i < 64;i++) {
m_plQuant[i] = LONG(delta[i]) << preshift;
m_plInvQuant[i] = LONG(FLOAT(1L << QUANTIZER_BITS) / delta[i] + 0.5);
}
}
///
/// IDCT::TransformBlock
// Run the DCT on a 8x8 block on the input data, giving the output table.
template<int preshift,typename T,bool deadzone,bool optimize>
void IDCT<preshift,T,deadzone,optimize>::TransformBlock(const LONG *source,LONG *target,LONG dcoffset)
{
LONG *dpend,*dp;
const LONG *qp = m_plInvQuant;
int band = 0;
//
// Adjust the DC offset to the number of fractional bits.
dcoffset <<= preshift + 3 + 3 + INTERMEDIATE_BITS;
// three additional bits because we still need to divide by 8.
//
// Pass over columns.
for(dp = target,dpend = target + 8;dp < dpend;dp++,source++) {
T tmp0 = source[0 << 3] + source[7 << 3];
T tmp1 = source[1 << 3] + source[6 << 3];
T tmp2 = source[2 << 3] + source[5 << 3];
T tmp3 = source[3 << 3] + source[4 << 3];
T tmp10 = tmp0 + tmp3;
T tmp12 = tmp0 - tmp3;
T tmp11 = tmp1 + tmp2;
T tmp13 = tmp1 - tmp2;
tmp0 = source[0 << 3] - source[7 << 3];
tmp1 = source[1 << 3] - source[6 << 3];
tmp2 = source[2 << 3] - source[5 << 3];
tmp3 = source[3 << 3] - source[4 << 3];
// complete DC and middle band.
dp[0 << 3] = (tmp10 + tmp11) << INTERMEDIATE_BITS;
dp[4 << 3] = (tmp10 - tmp11) << INTERMEDIATE_BITS;
FIXED z1 = (tmp12 + tmp13) * TO_FIX(0.541196100);
// complete bands 2 and 6
dp[2 << 3] = FIXED_TO_INTERMEDIATE(z1 + tmp12 * TO_FIX(0.765366865));
dp[6 << 3] = FIXED_TO_INTERMEDIATE(z1 + tmp13 *-TO_FIX(1.847759065));
tmp10 = tmp0 + tmp3;
tmp11 = tmp1 + tmp2;
tmp12 = tmp0 + tmp2;
tmp13 = tmp1 + tmp3;
z1 = (tmp12 + tmp13) * TO_FIX(1.175875602);
FIXED ttmp0 = tmp0 * TO_FIX(1.501321110);
FIXED ttmp1 = tmp1 * TO_FIX(3.072711026);
FIXED ttmp2 = tmp2 * TO_FIX(2.053119869);
FIXED ttmp3 = tmp3 * TO_FIX(0.298631336);
FIXED ttmp10 = tmp10 *-TO_FIX(0.899976223);
FIXED ttmp11 = tmp11 *-TO_FIX(2.562915447);
FIXED ttmp12 = tmp12 *-TO_FIX(0.390180644) + z1;
FIXED ttmp13 = tmp13 *-TO_FIX(1.961570560) + z1;
dp[1 << 3] = FIXED_TO_INTERMEDIATE(ttmp0 + ttmp10 + ttmp12);
dp[3 << 3] = FIXED_TO_INTERMEDIATE(ttmp1 + ttmp11 + ttmp13);
dp[5 << 3] = FIXED_TO_INTERMEDIATE(ttmp2 + ttmp11 + ttmp12);
dp[7 << 3] = FIXED_TO_INTERMEDIATE(ttmp3 + ttmp10 + ttmp13);
}
//
// Pass over rows and quantize.
for(dp = target,dpend = target + (8 << 3);dp < dpend;dp += 8,qp += 8) {
INTER tmp0 = dp[0] + dp[7];
INTER tmp1 = dp[1] + dp[6];
INTER tmp2 = dp[2] + dp[5];
INTER tmp3 = dp[3] + dp[4];
INTER tmp10 = tmp0 + tmp3;
INTER tmp12 = tmp0 - tmp3;
INTER tmp11 = tmp1 + tmp2;
INTER tmp13 = tmp1 - tmp2;
tmp0 = dp[0] - dp[7];
tmp1 = dp[1] - dp[6];
tmp2 = dp[2] - dp[5];
tmp3 = dp[3] - dp[4];
// complete DC and middle band.
dp[0] = Quantize((tmp10 + tmp11 - dcoffset) << FIX_BITS,qp[0],band);
dp[4] = Quantize((tmp10 - tmp11) << FIX_BITS ,qp[4],band+4);
INTER_FIXED z1 = (tmp12 + tmp13) * TO_FIX(0.541196100);
// complete bands 2 and 6
dp[2] = Quantize(z1 + tmp12 * TO_FIX(0.765366865),qp[2],band+2);
dp[6] = Quantize(z1 + tmp13 *-TO_FIX(1.847759065),qp[6],band+6);
tmp10 = tmp0 + tmp3;
tmp11 = tmp1 + tmp2;
tmp12 = tmp0 + tmp2;
tmp13 = tmp1 + tmp3;
z1 = (tmp12 + tmp13) * TO_FIX(1.175875602);
INTER_FIXED ttmp0 = tmp0 * TO_FIX(1.501321110);
INTER_FIXED ttmp1 = tmp1 * TO_FIX(3.072711026);
INTER_FIXED ttmp2 = tmp2 * TO_FIX(2.053119869);
INTER_FIXED ttmp3 = tmp3 * TO_FIX(0.298631336);
INTER_FIXED ttmp10 = tmp10 *-TO_FIX(0.899976223);
INTER_FIXED ttmp11 = tmp11 *-TO_FIX(2.562915447);
INTER_FIXED ttmp12 = tmp12 *-TO_FIX(0.390180644) + z1;
INTER_FIXED ttmp13 = tmp13 *-TO_FIX(1.961570560) + z1;
dp[1] = Quantize(ttmp0 + ttmp10 + ttmp12,qp[1],band+1);
dp[3] = Quantize(ttmp1 + ttmp11 + ttmp13,qp[3],band+3);
dp[5] = Quantize(ttmp2 + ttmp11 + ttmp12,qp[5],band+5);
dp[7] = Quantize(ttmp3 + ttmp10 + ttmp13,qp[7],band+7);
dcoffset = 0;
band += 8;
}
}
///
/// IDCT::InverseTransformBlock
// Run the inverse DCT on an 8x8 block reconstructing the data.
template<int preshift,typename T,bool deadzone,bool optimize>
void IDCT<preshift,T,deadzone,optimize>::InverseTransformBlock(LONG *target,const LONG *source,
LONG dcoffset)
{
LONG *dptr,*dend;
const LONG *qnt = m_plQuant;
dcoffset <<= preshift + 3;
if (source) {
for(dptr = target,dend = target + (8 << 3);dptr < dend;dptr +=8,source += 8,qnt += 8) {
// Even part.
T tz2 = source[2] * qnt[2];
T tz3 = source[6] * qnt[6];
FIXED z1 = (tz2 + tz3) * TO_FIX(0.541196100);
FIXED tmp2 = z1 + tz3 * -TO_FIX(1.847759065);
FIXED tmp3 = z1 + tz2 * TO_FIX(0.765366865);
tz2 = source[0] * qnt[0] + dcoffset;
tz3 = source[4] * qnt[4];
FIXED tmp0 = (tz2 + tz3) << FIX_BITS;
FIXED tmp1 = (tz2 - tz3) << FIX_BITS;
FIXED tmp10 = tmp0 + tmp3;
FIXED tmp13 = tmp0 - tmp3;
FIXED tmp11 = tmp1 + tmp2;
FIXED tmp12 = tmp1 - tmp2;
// Odd part.
T ttmp0 = source[7] * qnt[7];
T ttmp1 = source[5] * qnt[5];
T ttmp2 = source[3] * qnt[3];
T ttmp3 = source[1] * qnt[1];
T tz1 = ttmp0 + ttmp3;
tz2 = ttmp1 + ttmp2;
tz3 = ttmp0 + ttmp2;
T tz4 = ttmp1 + ttmp3;
FIXED z5 = (tz3 + tz4) * TO_FIX(1.175875602);
tmp0 = ttmp0 * TO_FIX(0.298631336);
tmp1 = ttmp1 * TO_FIX(2.053119869);
tmp2 = ttmp2 * TO_FIX(3.072711026);
tmp3 = ttmp3 * TO_FIX(1.501321110);
z1 = tz1 *-TO_FIX(0.899976223);
FIXED z2 = tz2 *-TO_FIX(2.562915447);
FIXED z3 = tz3 *-TO_FIX(1.961570560) + z5;
FIXED z4 = tz4 *-TO_FIX(0.390180644) + z5;
tmp0 += z1 + z3;
tmp1 += z2 + z4;
tmp2 += z2 + z3;
tmp3 += z1 + z4;
dptr[0] = FIXED_TO_INTERMEDIATE(tmp10 + tmp3);
dptr[7] = FIXED_TO_INTERMEDIATE(tmp10 - tmp3);
dptr[1] = FIXED_TO_INTERMEDIATE(tmp11 + tmp2);
dptr[6] = FIXED_TO_INTERMEDIATE(tmp11 - tmp2);
dptr[2] = FIXED_TO_INTERMEDIATE(tmp12 + tmp1);
dptr[5] = FIXED_TO_INTERMEDIATE(tmp12 - tmp1);
dptr[3] = FIXED_TO_INTERMEDIATE(tmp13 + tmp0);
dptr[4] = FIXED_TO_INTERMEDIATE(tmp13 - tmp0);
dcoffset = 0;
}
// After transforming over the columns, now transform over the rows.
for(dptr = target,dend = target + 8;dptr < dend;dptr++) {
INTER tz2 = dptr[2 << 3];
INTER tz3 = dptr[6 << 3];
INTER_FIXED z1 = (tz2 + tz3) * TO_FIX(0.541196100);
INTER_FIXED tmp2 = z1 + tz3 * -TO_FIX(1.847759065);
INTER_FIXED tmp3 = z1 + tz2 * TO_FIX(0.765366865);
INTER_FIXED tmp0 = (dptr[0 << 3] + dptr[4 << 3]) << FIX_BITS;
INTER_FIXED tmp1 = (dptr[0 << 3] - dptr[4 << 3]) << FIX_BITS;
INTER_FIXED tmp10 = tmp0 + tmp3;
INTER_FIXED tmp13 = tmp0 - tmp3;
INTER_FIXED tmp11 = tmp1 + tmp2;
INTER_FIXED tmp12 = tmp1 - tmp2;
// Odd parts.
INTER ttmp0 = dptr[7 << 3];
INTER ttmp1 = dptr[5 << 3];
INTER ttmp2 = dptr[3 << 3];
INTER ttmp3 = dptr[1 << 3];
INTER tz1 = ttmp0 + ttmp3;
tz2 = ttmp1 + ttmp2;
tz3 = ttmp0 + ttmp2;
INTER tz4 = ttmp1 + ttmp3;
INTER_FIXED z5 = (tz3 + tz4) * TO_FIX(1.175875602);
tmp0 = ttmp0 * TO_FIX(0.298631336);
tmp1 = ttmp1 * TO_FIX(2.053119869);
tmp2 = ttmp2 * TO_FIX(3.072711026);
tmp3 = ttmp3 * TO_FIX(1.501321110);
z1 = tz1 *-TO_FIX(0.899976223);
INTER_FIXED z2 = tz2 *-TO_FIX(2.562915447);
INTER_FIXED z3 = tz3 *-TO_FIX(1.961570560) + z5;
INTER_FIXED z4 = tz4 *-TO_FIX(0.390180644) + z5;
tmp0 += z1 + z3;
tmp1 += z2 + z4;
tmp2 += z2 + z3;
tmp3 += z1 + z4;
dptr[0 << 3] = INTER_FIXED_TO_INT(tmp10 + tmp3);
dptr[7 << 3] = INTER_FIXED_TO_INT(tmp10 - tmp3);
dptr[1 << 3] = INTER_FIXED_TO_INT(tmp11 + tmp2);
dptr[6 << 3] = INTER_FIXED_TO_INT(tmp11 - tmp2);
dptr[2 << 3] = INTER_FIXED_TO_INT(tmp12 + tmp1);
dptr[5 << 3] = INTER_FIXED_TO_INT(tmp12 - tmp1);
dptr[3 << 3] = INTER_FIXED_TO_INT(tmp13 + tmp0);
dptr[4 << 3] = INTER_FIXED_TO_INT(tmp13 - tmp0);
}
} else {
memset(target,0,sizeof(LONG) * 64);
}
}
///
/// IDCT::EstimateCriticalSlope
// Estimate a critical slope (lambda) from the unquantized data.
// Or to be precise, estimate lambda/delta^2, the constant in front of
// delta^2.
template<int preshift,typename T,bool deadzone,bool optimize>
DOUBLE IDCT<preshift,T,deadzone,optimize>::EstimateCriticalSlope(void)
{
#ifdef ESTIMATE_FROM_ENERGY
int i;
double energy = 0.0;
const double s1 = pow(2.0,14.75);
const double s2 = pow(2.0,16.5);
const double scale = double(1L << preshift) / 8.0;
assert(optimize);
for(i = 1;i < 63;i++) {
double val = m_lTransform[i] / scale;
energy += val * val;
}
energy /= 63.0;
return (s1 / (s2 + energy));
#else
return 0.25;
#endif
}
///
/// Instanciate the classes
template class IDCT<0,LONG,false,false>;
template class IDCT<1,LONG,false,false>; // For the RCT output
template class IDCT<ColorTrafo::COLOR_BITS,LONG,false,false>;
template class IDCT<ColorTrafo::COLOR_BITS,QUAD,false,false>;
template class IDCT<0,LONG,true,false>;
template class IDCT<1,LONG,true,false>; // For the RCT output
template class IDCT<ColorTrafo::COLOR_BITS,LONG,true,false>;
template class IDCT<ColorTrafo::COLOR_BITS,QUAD,true,false>;
template class IDCT<0,LONG,false,true>;
template class IDCT<1,LONG,false,true>; // For the RCT output
template class IDCT<ColorTrafo::COLOR_BITS,LONG,false,true>;
template class IDCT<ColorTrafo::COLOR_BITS,QUAD,false,true>;
template class IDCT<0,LONG,true,true>;
template class IDCT<1,LONG,true,true>; // For the RCT output
template class IDCT<ColorTrafo::COLOR_BITS,LONG,true,true>;
template class IDCT<ColorTrafo::COLOR_BITS,QUAD,true,true>;
///
|
