1 /*
2 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
3 *
4 * This code is free software; you can redistribute it and/or modify it
5 * under the terms of the GNU General Public License version 2 only, as
6 * published by the Free Software Foundation. Oracle designates this
7 * particular file as subject to the "Classpath" exception as provided
8 * by Oracle in the LICENSE file that accompanied this code.
9 *
10 * This code is distributed in the hope that it will be useful, but WITHOUT
11 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
12 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
13 * version 2 for more details (a copy is included in the LICENSE file that
14 * accompanied this code).
15 *
16 * You should have received a copy of the GNU General Public License version
17 * 2 along with this work; if not, write to the Free Software Foundation,
18 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
19 *
20 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
21 * or visit www.oracle.com if you need additional information or have any
22 * questions.
23 */
24
25 // This file is available under and governed by the GNU General Public
26 // License version 2 only, as published by the Free Software Foundation.
27 // However, the following notice accompanied the original version of this
28 // file:
29 //
30 //---------------------------------------------------------------------------------
31 //
32 // Little Color Management System
33 // Copyright (c) 1998-2020 Marti Maria Saguer
34 //
35 // Permission is hereby granted, free of charge, to any person obtaining
36 // a copy of this software and associated documentation files (the "Software"),
37 // to deal in the Software without restriction, including without limitation
38 // the rights to use, copy, modify, merge, publish, distribute, sublicense,
39 // and/or sell copies of the Software, and to permit persons to whom the Software
40 // is furnished to do so, subject to the following conditions:
41 //
42 // The above copyright notice and this permission notice shall be included in
43 // all copies or substantial portions of the Software.
44 //
45 // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
46 // EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
47 // THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
48 // NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
49 // LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
50 // OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
51 // WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
52 //
53 //---------------------------------------------------------------------------------
54 //
55 #include "lcms2_internal.h"
56
57 // Tone curves are powerful constructs that can contain curves specified in diverse ways.
58 // The curve is stored in segments, where each segment can be sampled or specified by parameters.
59 // a 16.bit simplification of the *whole* curve is kept for optimization purposes. For float operation,
60 // each segment is evaluated separately. Plug-ins may be used to define new parametric schemes,
61 // each plug-in may define up to MAX_TYPES_IN_LCMS_PLUGIN functions types. For defining a function,
62 // the plug-in should provide the type id, how many parameters each type has, and a pointer to
63 // a procedure that evaluates the function. In the case of reverse evaluation, the evaluator will
64 // be called with the type id as a negative value, and a sampled version of the reversed curve
65 // will be built.
66
67 // ----------------------------------------------------------------- Implementation
68 // Maxim number of nodes
69 #define MAX_NODES_IN_CURVE 4097
70 #define MINUS_INF (-1E22F)
71 #define PLUS_INF (+1E22F)
72
73 // The list of supported parametric curves
74 typedef struct _cmsParametricCurvesCollection_st {
75
76 cmsUInt32Number nFunctions; // Number of supported functions in this chunk
77 cmsInt32Number FunctionTypes[MAX_TYPES_IN_LCMS_PLUGIN]; // The identification types
78 cmsUInt32Number ParameterCount[MAX_TYPES_IN_LCMS_PLUGIN]; // Number of parameters for each function
79
80 cmsParametricCurveEvaluator Evaluator; // The evaluator
81
82 struct _cmsParametricCurvesCollection_st* Next; // Next in list
83
84 } _cmsParametricCurvesCollection;
85
86 // This is the default (built-in) evaluator
87 static cmsFloat64Number DefaultEvalParametricFn(cmsInt32Number Type, const cmsFloat64Number Params[], cmsFloat64Number R);
88
89 // The built-in list
90 static _cmsParametricCurvesCollection DefaultCurves = {
91 9, // # of curve types
92 { 1, 2, 3, 4, 5, 6, 7, 8, 108 }, // Parametric curve ID
93 { 1, 3, 4, 5, 7, 4, 5, 5, 1 }, // Parameters by type
94 DefaultEvalParametricFn, // Evaluator
95 NULL // Next in chain
96 };
97
98 // Duplicates the zone of memory used by the plug-in in the new context
99 static
100 void DupPluginCurvesList(struct _cmsContext_struct* ctx,
101 const struct _cmsContext_struct* src)
102 {
103 _cmsCurvesPluginChunkType newHead = { NULL };
104 _cmsParametricCurvesCollection* entry;
105 _cmsParametricCurvesCollection* Anterior = NULL;
106 _cmsCurvesPluginChunkType* head = (_cmsCurvesPluginChunkType*) src->chunks[CurvesPlugin];
107
108 _cmsAssert(head != NULL);
109
110 // Walk the list copying all nodes
111 for (entry = head->ParametricCurves;
112 entry != NULL;
113 entry = entry ->Next) {
114
115 _cmsParametricCurvesCollection *newEntry = ( _cmsParametricCurvesCollection *) _cmsSubAllocDup(ctx ->MemPool, entry, sizeof(_cmsParametricCurvesCollection));
116
117 if (newEntry == NULL)
118 return;
119
120 // We want to keep the linked list order, so this is a little bit tricky
121 newEntry -> Next = NULL;
122 if (Anterior)
123 Anterior -> Next = newEntry;
124
125 Anterior = newEntry;
126
127 if (newHead.ParametricCurves == NULL)
128 newHead.ParametricCurves = newEntry;
129 }
130
131 ctx ->chunks[CurvesPlugin] = _cmsSubAllocDup(ctx->MemPool, &newHead, sizeof(_cmsCurvesPluginChunkType));
132 }
133
134 // The allocator have to follow the chain
135 void _cmsAllocCurvesPluginChunk(struct _cmsContext_struct* ctx,
136 const struct _cmsContext_struct* src)
137 {
138 _cmsAssert(ctx != NULL);
139
140 if (src != NULL) {
141
142 // Copy all linked list
143 DupPluginCurvesList(ctx, src);
144 }
145 else {
146 static _cmsCurvesPluginChunkType CurvesPluginChunk = { NULL };
147 ctx ->chunks[CurvesPlugin] = _cmsSubAllocDup(ctx ->MemPool, &CurvesPluginChunk, sizeof(_cmsCurvesPluginChunkType));
148 }
149 }
150
151
152 // The linked list head
153 _cmsCurvesPluginChunkType _cmsCurvesPluginChunk = { NULL };
154
155 // As a way to install new parametric curves
156 cmsBool _cmsRegisterParametricCurvesPlugin(cmsContext ContextID, cmsPluginBase* Data)
157 {
158 _cmsCurvesPluginChunkType* ctx = ( _cmsCurvesPluginChunkType*) _cmsContextGetClientChunk(ContextID, CurvesPlugin);
159 cmsPluginParametricCurves* Plugin = (cmsPluginParametricCurves*) Data;
160 _cmsParametricCurvesCollection* fl;
161
162 if (Data == NULL) {
163
164 ctx -> ParametricCurves = NULL;
165 return TRUE;
166 }
167
168 fl = (_cmsParametricCurvesCollection*) _cmsPluginMalloc(ContextID, sizeof(_cmsParametricCurvesCollection));
169 if (fl == NULL) return FALSE;
170
171 // Copy the parameters
172 fl ->Evaluator = Plugin ->Evaluator;
173 fl ->nFunctions = Plugin ->nFunctions;
174
175 // Make sure no mem overwrites
176 if (fl ->nFunctions > MAX_TYPES_IN_LCMS_PLUGIN)
177 fl ->nFunctions = MAX_TYPES_IN_LCMS_PLUGIN;
178
179 // Copy the data
180 memmove(fl->FunctionTypes, Plugin ->FunctionTypes, fl->nFunctions * sizeof(cmsUInt32Number));
181 memmove(fl->ParameterCount, Plugin ->ParameterCount, fl->nFunctions * sizeof(cmsUInt32Number));
182
183 // Keep linked list
184 fl ->Next = ctx->ParametricCurves;
185 ctx->ParametricCurves = fl;
186
187 // All is ok
188 return TRUE;
189 }
190
191
192 // Search in type list, return position or -1 if not found
193 static
194 int IsInSet(int Type, _cmsParametricCurvesCollection* c)
195 {
196 int i;
197
198 for (i=0; i < (int) c ->nFunctions; i++)
199 if (abs(Type) == c ->FunctionTypes[i]) return i;
200
201 return -1;
202 }
203
204
205 // Search for the collection which contains a specific type
206 static
207 _cmsParametricCurvesCollection *GetParametricCurveByType(cmsContext ContextID, int Type, int* index)
208 {
209 _cmsParametricCurvesCollection* c;
210 int Position;
211 _cmsCurvesPluginChunkType* ctx = ( _cmsCurvesPluginChunkType*) _cmsContextGetClientChunk(ContextID, CurvesPlugin);
212
213 for (c = ctx->ParametricCurves; c != NULL; c = c ->Next) {
214
215 Position = IsInSet(Type, c);
216
217 if (Position != -1) {
218 if (index != NULL)
219 *index = Position;
220 return c;
221 }
222 }
223 // If none found, revert for defaults
224 for (c = &DefaultCurves; c != NULL; c = c ->Next) {
225
226 Position = IsInSet(Type, c);
227
228 if (Position != -1) {
229 if (index != NULL)
230 *index = Position;
231 return c;
232 }
233 }
234
235 return NULL;
236 }
237
238 // Low level allocate, which takes care of memory details. nEntries may be zero, and in this case
239 // no optimation curve is computed. nSegments may also be zero in the inverse case, where only the
240 // optimization curve is given. Both features simultaneously is an error
241 static
242 cmsToneCurve* AllocateToneCurveStruct(cmsContext ContextID, cmsUInt32Number nEntries,
243 cmsUInt32Number nSegments, const cmsCurveSegment* Segments,
244 const cmsUInt16Number* Values)
245 {
246 cmsToneCurve* p;
247 cmsUInt32Number i;
248
249 // We allow huge tables, which are then restricted for smoothing operations
250 if (nEntries > 65530) {
251 cmsSignalError(ContextID, cmsERROR_RANGE, "Couldn't create tone curve of more than 65530 entries");
252 return NULL;
253 }
254
255 if (nEntries == 0 && nSegments == 0) {
256 cmsSignalError(ContextID, cmsERROR_RANGE, "Couldn't create tone curve with zero segments and no table");
257 return NULL;
258 }
259
260 // Allocate all required pointers, etc.
261 p = (cmsToneCurve*) _cmsMallocZero(ContextID, sizeof(cmsToneCurve));
262 if (!p) return NULL;
263
264 // In this case, there are no segments
265 if (nSegments == 0) {
266 p ->Segments = NULL;
267 p ->Evals = NULL;
268 }
269 else {
270 p ->Segments = (cmsCurveSegment*) _cmsCalloc(ContextID, nSegments, sizeof(cmsCurveSegment));
271 if (p ->Segments == NULL) goto Error;
272
273 p ->Evals = (cmsParametricCurveEvaluator*) _cmsCalloc(ContextID, nSegments, sizeof(cmsParametricCurveEvaluator));
274 if (p ->Evals == NULL) goto Error;
275 }
276
277 p -> nSegments = nSegments;
278
279 // This 16-bit table contains a limited precision representation of the whole curve and is kept for
280 // increasing xput on certain operations.
281 if (nEntries == 0) {
282 p ->Table16 = NULL;
283 }
284 else {
285 p ->Table16 = (cmsUInt16Number*) _cmsCalloc(ContextID, nEntries, sizeof(cmsUInt16Number));
286 if (p ->Table16 == NULL) goto Error;
287 }
288
289 p -> nEntries = nEntries;
290
291 // Initialize members if requested
292 if (Values != NULL && (nEntries > 0)) {
293
294 for (i=0; i < nEntries; i++)
295 p ->Table16[i] = Values[i];
296 }
297
298 // Initialize the segments stuff. The evaluator for each segment is located and a pointer to it
299 // is placed in advance to maximize performance.
300 if (Segments != NULL && (nSegments > 0)) {
301
302 _cmsParametricCurvesCollection *c;
303
304 p ->SegInterp = (cmsInterpParams**) _cmsCalloc(ContextID, nSegments, sizeof(cmsInterpParams*));
305 if (p ->SegInterp == NULL) goto Error;
306
307 for (i=0; i < nSegments; i++) {
308
309 // Type 0 is a special marker for table-based curves
310 if (Segments[i].Type == 0)
311 p ->SegInterp[i] = _cmsComputeInterpParams(ContextID, Segments[i].nGridPoints, 1, 1, NULL, CMS_LERP_FLAGS_FLOAT);
312
313 memmove(&p ->Segments[i], &Segments[i], sizeof(cmsCurveSegment));
314
315 if (Segments[i].Type == 0 && Segments[i].SampledPoints != NULL)
316 p ->Segments[i].SampledPoints = (cmsFloat32Number*) _cmsDupMem(ContextID, Segments[i].SampledPoints, sizeof(cmsFloat32Number) * Segments[i].nGridPoints);
317 else
318 p ->Segments[i].SampledPoints = NULL;
319
320
321 c = GetParametricCurveByType(ContextID, Segments[i].Type, NULL);
322 if (c != NULL)
323 p ->Evals[i] = c ->Evaluator;
324 }
325 }
326
327 p ->InterpParams = _cmsComputeInterpParams(ContextID, p ->nEntries, 1, 1, p->Table16, CMS_LERP_FLAGS_16BITS);
328 if (p->InterpParams != NULL)
329 return p;
330
331 Error:
332 if (p -> SegInterp) _cmsFree(ContextID, p -> SegInterp);
333 if (p -> Segments) _cmsFree(ContextID, p -> Segments);
334 if (p -> Evals) _cmsFree(ContextID, p -> Evals);
335 if (p ->Table16) _cmsFree(ContextID, p ->Table16);
336 _cmsFree(ContextID, p);
337 return NULL;
338 }
339
340
341 // Parametric Fn using floating point
342 static
343 cmsFloat64Number DefaultEvalParametricFn(cmsInt32Number Type, const cmsFloat64Number Params[], cmsFloat64Number R)
344 {
345 cmsFloat64Number e, Val, disc;
346
347 switch (Type) {
348
349 // X = Y ^ Gamma
350 case 1:
351 if (R < 0) {
352
353 if (fabs(Params[0] - 1.0) < MATRIX_DET_TOLERANCE)
354 Val = R;
355 else
356 Val = 0;
357 }
358 else
359 Val = pow(R, Params[0]);
360 break;
361
362 // Type 1 Reversed: X = Y ^1/gamma
363 case -1:
364 if (R < 0) {
365
366 if (fabs(Params[0] - 1.0) < MATRIX_DET_TOLERANCE)
367 Val = R;
368 else
369 Val = 0;
370 }
371 else
372 {
373 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE)
374 Val = PLUS_INF;
375 else
376 Val = pow(R, 1 / Params[0]);
377 }
378 break;
379
380 // CIE 122-1966
381 // Y = (aX + b)^Gamma | X >= -b/a
382 // Y = 0 | else
383 case 2:
384 {
385
386 if (fabs(Params[1]) < MATRIX_DET_TOLERANCE)
387 {
388 Val = 0;
389 }
390 else
391 {
392 disc = -Params[2] / Params[1];
393
394 if (R >= disc) {
395
396 e = Params[1] * R + Params[2];
397
398 if (e > 0)
399 Val = pow(e, Params[0]);
400 else
401 Val = 0;
402 }
403 else
404 Val = 0;
405 }
406 }
407 break;
408
409 // Type 2 Reversed
410 // X = (Y ^1/g - b) / a
411 case -2:
412 {
413 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE ||
414 fabs(Params[1]) < MATRIX_DET_TOLERANCE)
415 {
416 Val = 0;
417 }
418 else
419 {
420 if (R < 0)
421 Val = 0;
422 else
423 Val = (pow(R, 1.0 / Params[0]) - Params[2]) / Params[1];
424
425 if (Val < 0)
426 Val = 0;
427 }
428 }
429 break;
430
431
432 // IEC 61966-3
433 // Y = (aX + b)^Gamma | X <= -b/a
434 // Y = c | else
435 case 3:
436 {
437 if (fabs(Params[1]) < MATRIX_DET_TOLERANCE)
438 {
439 Val = 0;
440 }
441 else
442 {
443 disc = -Params[2] / Params[1];
444 if (disc < 0)
445 disc = 0;
446
447 if (R >= disc) {
448
449 e = Params[1] * R + Params[2];
450
451 if (e > 0)
452 Val = pow(e, Params[0]) + Params[3];
453 else
454 Val = 0;
455 }
456 else
457 Val = Params[3];
458 }
459 }
460 break;
461
462
463 // Type 3 reversed
464 // X=((Y-c)^1/g - b)/a | (Y>=c)
465 // X=-b/a | (Y<c)
466 case -3:
467 {
468 if (fabs(Params[1]) < MATRIX_DET_TOLERANCE)
469 {
470 Val = 0;
471 }
472 else
473 {
474 if (R >= Params[3]) {
475
476 e = R - Params[3];
477
478 if (e > 0)
479 Val = (pow(e, 1 / Params[0]) - Params[2]) / Params[1];
480 else
481 Val = 0;
482 }
483 else {
484 Val = -Params[2] / Params[1];
485 }
486 }
487 }
488 break;
489
490
491 // IEC 61966-2.1 (sRGB)
492 // Y = (aX + b)^Gamma | X >= d
493 // Y = cX | X < d
494 case 4:
495 if (R >= Params[4]) {
496
497 e = Params[1]*R + Params[2];
498
499 if (e > 0)
500 Val = pow(e, Params[0]);
501 else
502 Val = 0;
503 }
504 else
505 Val = R * Params[3];
506 break;
507
508 // Type 4 reversed
509 // X=((Y^1/g-b)/a) | Y >= (ad+b)^g
510 // X=Y/c | Y< (ad+b)^g
511 case -4:
512 {
513 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE ||
514 fabs(Params[1]) < MATRIX_DET_TOLERANCE ||
515 fabs(Params[3]) < MATRIX_DET_TOLERANCE)
516 {
517 Val = 0;
518 }
519 else
520 {
521 e = Params[1] * Params[4] + Params[2];
522 if (e < 0)
523 disc = 0;
524 else
525 disc = pow(e, Params[0]);
526
527 if (R >= disc) {
528
529 Val = (pow(R, 1.0 / Params[0]) - Params[2]) / Params[1];
530 }
531 else {
532 Val = R / Params[3];
533 }
534 }
535 }
536 break;
537
538
539 // Y = (aX + b)^Gamma + e | X >= d
540 // Y = cX + f | X < d
541 case 5:
542 if (R >= Params[4]) {
543
544 e = Params[1]*R + Params[2];
545
546 if (e > 0)
547 Val = pow(e, Params[0]) + Params[5];
548 else
549 Val = Params[5];
550 }
551 else
552 Val = R*Params[3] + Params[6];
553 break;
554
555
556 // Reversed type 5
557 // X=((Y-e)1/g-b)/a | Y >=(ad+b)^g+e), cd+f
558 // X=(Y-f)/c | else
559 case -5:
560 {
561 if (fabs(Params[1]) < MATRIX_DET_TOLERANCE ||
562 fabs(Params[3]) < MATRIX_DET_TOLERANCE)
563 {
564 Val = 0;
565 }
566 else
567 {
568 disc = Params[3] * Params[4] + Params[6];
569 if (R >= disc) {
570
571 e = R - Params[5];
572 if (e < 0)
573 Val = 0;
574 else
575 Val = (pow(e, 1.0 / Params[0]) - Params[2]) / Params[1];
576 }
577 else {
578 Val = (R - Params[6]) / Params[3];
579 }
580 }
581 }
582 break;
583
584
585 // Types 6,7,8 comes from segmented curves as described in ICCSpecRevision_02_11_06_Float.pdf
586 // Type 6 is basically identical to type 5 without d
587
588 // Y = (a * X + b) ^ Gamma + c
589 case 6:
590 e = Params[1]*R + Params[2];
591
592 if (e < 0)
593 Val = Params[3];
594 else
595 Val = pow(e, Params[0]) + Params[3];
596 break;
597
598 // ((Y - c) ^1/Gamma - b) / a
599 case -6:
600 {
601 if (fabs(Params[1]) < MATRIX_DET_TOLERANCE)
602 {
603 Val = 0;
604 }
605 else
606 {
607 e = R - Params[3];
608 if (e < 0)
609 Val = 0;
610 else
611 Val = (pow(e, 1.0 / Params[0]) - Params[2]) / Params[1];
612 }
613 }
614 break;
615
616
617 // Y = a * log (b * X^Gamma + c) + d
618 case 7:
619
620 e = Params[2] * pow(R, Params[0]) + Params[3];
621 if (e <= 0)
622 Val = Params[4];
623 else
624 Val = Params[1]*log10(e) + Params[4];
625 break;
626
627 // (Y - d) / a = log(b * X ^Gamma + c)
628 // pow(10, (Y-d) / a) = b * X ^Gamma + c
629 // pow((pow(10, (Y-d) / a) - c) / b, 1/g) = X
630 case -7:
631 {
632 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE ||
633 fabs(Params[1]) < MATRIX_DET_TOLERANCE ||
634 fabs(Params[2]) < MATRIX_DET_TOLERANCE)
635 {
636 Val = 0;
637 }
638 else
639 {
640 Val = pow((pow(10.0, (R - Params[4]) / Params[1]) - Params[3]) / Params[2], 1.0 / Params[0]);
641 }
642 }
643 break;
644
645
646 //Y = a * b^(c*X+d) + e
647 case 8:
648 Val = (Params[0] * pow(Params[1], Params[2] * R + Params[3]) + Params[4]);
649 break;
650
651
652 // Y = (log((y-e) / a) / log(b) - d ) / c
653 // a=0, b=1, c=2, d=3, e=4,
654 case -8:
655
656 disc = R - Params[4];
657 if (disc < 0) Val = 0;
658 else
659 {
660 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE ||
661 fabs(Params[2]) < MATRIX_DET_TOLERANCE)
662 {
663 Val = 0;
664 }
665 else
666 {
667 Val = (log(disc / Params[0]) / log(Params[1]) - Params[3]) / Params[2];
668 }
669 }
670 break;
671
672 // S-Shaped: (1 - (1-x)^1/g)^1/g
673 case 108:
674 if (fabs(Params[0]) < MATRIX_DET_TOLERANCE)
675 Val = 0;
676 else
677 Val = pow(1.0 - pow(1 - R, 1/Params[0]), 1/Params[0]);
678 break;
679
680 // y = (1 - (1-x)^1/g)^1/g
681 // y^g = (1 - (1-x)^1/g)
682 // 1 - y^g = (1-x)^1/g
683 // (1 - y^g)^g = 1 - x
684 // 1 - (1 - y^g)^g
685 case -108:
686 Val = 1 - pow(1 - pow(R, Params[0]), Params[0]);
687 break;
688
689 default:
690 // Unsupported parametric curve. Should never reach here
691 return 0;
692 }
693
694 return Val;
695 }
696
697 // Evaluate a segmented function for a single value. Return -Inf if no valid segment found .
698 // If fn type is 0, perform an interpolation on the table
699 static
700 cmsFloat64Number EvalSegmentedFn(const cmsToneCurve *g, cmsFloat64Number R)
701 {
702 int i;
703 cmsFloat32Number Out32;
704 cmsFloat64Number Out;
705
706 for (i = (int) g->nSegments - 1; i >= 0; --i) {
707
708 // Check for domain
709 if ((R > g->Segments[i].x0) && (R <= g->Segments[i].x1)) {
710
711 // Type == 0 means segment is sampled
712 if (g->Segments[i].Type == 0) {
713
714 cmsFloat32Number R1 = (cmsFloat32Number)(R - g->Segments[i].x0) / (g->Segments[i].x1 - g->Segments[i].x0);
715
716 // Setup the table (TODO: clean that)
717 g->SegInterp[i]->Table = g->Segments[i].SampledPoints;
718
719 g->SegInterp[i]->Interpolation.LerpFloat(&R1, &Out32, g->SegInterp[i]);
720 Out = (cmsFloat64Number) Out32;
721
722 }
723 else {
724 Out = g->Evals[i](g->Segments[i].Type, g->Segments[i].Params, R);
725 }
726
727 if (isinf(Out))
728 return PLUS_INF;
729 else
730 {
731 if (isinf(-Out))
732 return MINUS_INF;
733 }
734
735 return Out;
736 }
737 }
738
739 return MINUS_INF;
740 }
741
742 // Access to estimated low-res table
743 cmsUInt32Number CMSEXPORT cmsGetToneCurveEstimatedTableEntries(const cmsToneCurve* t)
744 {
745 _cmsAssert(t != NULL);
746 return t ->nEntries;
747 }
748
749 const cmsUInt16Number* CMSEXPORT cmsGetToneCurveEstimatedTable(const cmsToneCurve* t)
750 {
751 _cmsAssert(t != NULL);
752 return t ->Table16;
753 }
754
755
756 // Create an empty gamma curve, by using tables. This specifies only the limited-precision part, and leaves the
757 // floating point description empty.
758 cmsToneCurve* CMSEXPORT cmsBuildTabulatedToneCurve16(cmsContext ContextID, cmsUInt32Number nEntries, const cmsUInt16Number Values[])
759 {
760 return AllocateToneCurveStruct(ContextID, nEntries, 0, NULL, Values);
761 }
762
763 static
764 cmsUInt32Number EntriesByGamma(cmsFloat64Number Gamma)
765 {
766 if (fabs(Gamma - 1.0) < 0.001) return 2;
767 return 4096;
768 }
769
770
771 // Create a segmented gamma, fill the table
772 cmsToneCurve* CMSEXPORT cmsBuildSegmentedToneCurve(cmsContext ContextID,
773 cmsUInt32Number nSegments, const cmsCurveSegment Segments[])
774 {
775 cmsUInt32Number i;
776 cmsFloat64Number R, Val;
777 cmsToneCurve* g;
778 cmsUInt32Number nGridPoints = 4096;
779
780 _cmsAssert(Segments != NULL);
781
782 // Optimizatin for identity curves.
783 if (nSegments == 1 && Segments[0].Type == 1) {
784
785 nGridPoints = EntriesByGamma(Segments[0].Params[0]);
786 }
787
788 g = AllocateToneCurveStruct(ContextID, nGridPoints, nSegments, Segments, NULL);
789 if (g == NULL) return NULL;
790
791 // Once we have the floating point version, we can approximate a 16 bit table of 4096 entries
792 // for performance reasons. This table would normally not be used except on 8/16 bits transforms.
793 for (i = 0; i < nGridPoints; i++) {
794
795 R = (cmsFloat64Number) i / (nGridPoints-1);
796
797 Val = EvalSegmentedFn(g, R);
798
799 // Round and saturate
800 g ->Table16[i] = _cmsQuickSaturateWord(Val * 65535.0);
801 }
802
803 return g;
804 }
805
806 // Use a segmented curve to store the floating point table
807 cmsToneCurve* CMSEXPORT cmsBuildTabulatedToneCurveFloat(cmsContext ContextID, cmsUInt32Number nEntries, const cmsFloat32Number values[])
808 {
809 cmsCurveSegment Seg[3];
810
811 // A segmented tone curve should have function segments in the first and last positions
812 // Initialize segmented curve part up to 0 to constant value = samples[0]
813 Seg[0].x0 = MINUS_INF;
814 Seg[0].x1 = 0;
815 Seg[0].Type = 6;
816
817 Seg[0].Params[0] = 1;
818 Seg[0].Params[1] = 0;
819 Seg[0].Params[2] = 0;
820 Seg[0].Params[3] = values[0];
821 Seg[0].Params[4] = 0;
822
823 // From zero to 1
824 Seg[1].x0 = 0;
825 Seg[1].x1 = 1.0;
826 Seg[1].Type = 0;
827
828 Seg[1].nGridPoints = nEntries;
829 Seg[1].SampledPoints = (cmsFloat32Number*) values;
830
831 // Final segment is constant = lastsample
832 Seg[2].x0 = 1.0;
833 Seg[2].x1 = PLUS_INF;
834 Seg[2].Type = 6;
835
836 Seg[2].Params[0] = 1;
837 Seg[2].Params[1] = 0;
838 Seg[2].Params[2] = 0;
839 Seg[2].Params[3] = values[nEntries-1];
840 Seg[2].Params[4] = 0;
841
842
843 return cmsBuildSegmentedToneCurve(ContextID, 3, Seg);
844 }
845
846 // Parametric curves
847 //
848 // Parameters goes as: Curve, a, b, c, d, e, f
849 // Type is the ICC type +1
850 // if type is negative, then the curve is analytically inverted
851 cmsToneCurve* CMSEXPORT cmsBuildParametricToneCurve(cmsContext ContextID, cmsInt32Number Type, const cmsFloat64Number Params[])
852 {
853 cmsCurveSegment Seg0;
854 int Pos = 0;
855 cmsUInt32Number size;
856 _cmsParametricCurvesCollection* c = GetParametricCurveByType(ContextID, Type, &Pos);
857
858 _cmsAssert(Params != NULL);
859
860 if (c == NULL) {
861 cmsSignalError(ContextID, cmsERROR_UNKNOWN_EXTENSION, "Invalid parametric curve type %d", Type);
862 return NULL;
863 }
864
865 memset(&Seg0, 0, sizeof(Seg0));
866
867 Seg0.x0 = MINUS_INF;
868 Seg0.x1 = PLUS_INF;
869 Seg0.Type = Type;
870
871 size = c->ParameterCount[Pos] * sizeof(cmsFloat64Number);
872 memmove(Seg0.Params, Params, size);
873
874 return cmsBuildSegmentedToneCurve(ContextID, 1, &Seg0);
875 }
876
877
878
879 // Build a gamma table based on gamma constant
880 cmsToneCurve* CMSEXPORT cmsBuildGamma(cmsContext ContextID, cmsFloat64Number Gamma)
881 {
882 return cmsBuildParametricToneCurve(ContextID, 1, &Gamma);
883 }
884
885
886 // Free all memory taken by the gamma curve
887 void CMSEXPORT cmsFreeToneCurve(cmsToneCurve* Curve)
888 {
889 cmsContext ContextID;
890
891 if (Curve == NULL) return;
892
893 ContextID = Curve ->InterpParams->ContextID;
894
895 _cmsFreeInterpParams(Curve ->InterpParams);
896
897 if (Curve -> Table16)
898 _cmsFree(ContextID, Curve ->Table16);
899
900 if (Curve ->Segments) {
901
902 cmsUInt32Number i;
903
904 for (i=0; i < Curve ->nSegments; i++) {
905
906 if (Curve ->Segments[i].SampledPoints) {
907 _cmsFree(ContextID, Curve ->Segments[i].SampledPoints);
908 }
909
910 if (Curve ->SegInterp[i] != 0)
911 _cmsFreeInterpParams(Curve->SegInterp[i]);
912 }
913
914 _cmsFree(ContextID, Curve ->Segments);
915 _cmsFree(ContextID, Curve ->SegInterp);
916 }
917
918 if (Curve -> Evals)
919 _cmsFree(ContextID, Curve -> Evals);
920
921 _cmsFree(ContextID, Curve);
922 }
923
924 // Utility function, free 3 gamma tables
925 void CMSEXPORT cmsFreeToneCurveTriple(cmsToneCurve* Curve[3])
926 {
927
928 _cmsAssert(Curve != NULL);
929
930 if (Curve[0] != NULL) cmsFreeToneCurve(Curve[0]);
931 if (Curve[1] != NULL) cmsFreeToneCurve(Curve[1]);
932 if (Curve[2] != NULL) cmsFreeToneCurve(Curve[2]);
933
934 Curve[0] = Curve[1] = Curve[2] = NULL;
935 }
936
937
938 // Duplicate a gamma table
939 cmsToneCurve* CMSEXPORT cmsDupToneCurve(const cmsToneCurve* In)
940 {
941 if (In == NULL) return NULL;
942
943 return AllocateToneCurveStruct(In ->InterpParams ->ContextID, In ->nEntries, In ->nSegments, In ->Segments, In ->Table16);
944 }
945
946 // Joins two curves for X and Y. Curves should be monotonic.
947 // We want to get
948 //
949 // y = Y^-1(X(t))
950 //
951 cmsToneCurve* CMSEXPORT cmsJoinToneCurve(cmsContext ContextID,
952 const cmsToneCurve* X,
953 const cmsToneCurve* Y, cmsUInt32Number nResultingPoints)
954 {
955 cmsToneCurve* out = NULL;
956 cmsToneCurve* Yreversed = NULL;
957 cmsFloat32Number t, x;
958 cmsFloat32Number* Res = NULL;
959 cmsUInt32Number i;
960
961
962 _cmsAssert(X != NULL);
963 _cmsAssert(Y != NULL);
964
965 Yreversed = cmsReverseToneCurveEx(nResultingPoints, Y);
966 if (Yreversed == NULL) goto Error;
967
968 Res = (cmsFloat32Number*) _cmsCalloc(ContextID, nResultingPoints, sizeof(cmsFloat32Number));
969 if (Res == NULL) goto Error;
970
971 //Iterate
972 for (i=0; i < nResultingPoints; i++) {
973
974 t = (cmsFloat32Number) i / (nResultingPoints-1);
975 x = cmsEvalToneCurveFloat(X, t);
976 Res[i] = cmsEvalToneCurveFloat(Yreversed, x);
977 }
978
979 // Allocate space for output
980 out = cmsBuildTabulatedToneCurveFloat(ContextID, nResultingPoints, Res);
981
982 Error:
983
984 if (Res != NULL) _cmsFree(ContextID, Res);
985 if (Yreversed != NULL) cmsFreeToneCurve(Yreversed);
986
987 return out;
988 }
989
990
991
992 // Get the surrounding nodes. This is tricky on non-monotonic tables
993 static
994 int GetInterval(cmsFloat64Number In, const cmsUInt16Number LutTable[], const struct _cms_interp_struc* p)
995 {
996 int i;
997 int y0, y1;
998
999 // A 1 point table is not allowed
1000 if (p -> Domain[0] < 1) return -1;
1001
1002 // Let's see if ascending or descending.
1003 if (LutTable[0] < LutTable[p ->Domain[0]]) {
1004
1005 // Table is overall ascending
1006 for (i = (int) p->Domain[0] - 1; i >= 0; --i) {
1007
1008 y0 = LutTable[i];
1009 y1 = LutTable[i+1];
1010
1011 if (y0 <= y1) { // Increasing
1012 if (In >= y0 && In <= y1) return i;
1013 }
1014 else
1015 if (y1 < y0) { // Decreasing
1016 if (In >= y1 && In <= y0) return i;
1017 }
1018 }
1019 }
1020 else {
1021 // Table is overall descending
1022 for (i=0; i < (int) p -> Domain[0]; i++) {
1023
1024 y0 = LutTable[i];
1025 y1 = LutTable[i+1];
1026
1027 if (y0 <= y1) { // Increasing
1028 if (In >= y0 && In <= y1) return i;
1029 }
1030 else
1031 if (y1 < y0) { // Decreasing
1032 if (In >= y1 && In <= y0) return i;
1033 }
1034 }
1035 }
1036
1037 return -1;
1038 }
1039
1040 // Reverse a gamma table
1041 cmsToneCurve* CMSEXPORT cmsReverseToneCurveEx(cmsUInt32Number nResultSamples, const cmsToneCurve* InCurve)
1042 {
1043 cmsToneCurve *out;
1044 cmsFloat64Number a = 0, b = 0, y, x1, y1, x2, y2;
1045 int i, j;
1046 int Ascending;
1047
1048 _cmsAssert(InCurve != NULL);
1049
1050 // Try to reverse it analytically whatever possible
1051
1052 if (InCurve ->nSegments == 1 && InCurve ->Segments[0].Type > 0 &&
1053 /* InCurve -> Segments[0].Type <= 5 */
1054 GetParametricCurveByType(InCurve ->InterpParams->ContextID, InCurve ->Segments[0].Type, NULL) != NULL) {
1055
1056 return cmsBuildParametricToneCurve(InCurve ->InterpParams->ContextID,
1057 -(InCurve -> Segments[0].Type),
1058 InCurve -> Segments[0].Params);
1059 }
1060
1061 // Nope, reverse the table.
1062 out = cmsBuildTabulatedToneCurve16(InCurve ->InterpParams->ContextID, nResultSamples, NULL);
1063 if (out == NULL)
1064 return NULL;
1065
1066 // We want to know if this is an ascending or descending table
1067 Ascending = !cmsIsToneCurveDescending(InCurve);
1068
1069 // Iterate across Y axis
1070 for (i=0; i < (int) nResultSamples; i++) {
1071
1072 y = (cmsFloat64Number) i * 65535.0 / (nResultSamples - 1);
1073
1074 // Find interval in which y is within.
1075 j = GetInterval(y, InCurve->Table16, InCurve->InterpParams);
1076 if (j >= 0) {
1077
1078
1079 // Get limits of interval
1080 x1 = InCurve ->Table16[j];
1081 x2 = InCurve ->Table16[j+1];
1082
1083 y1 = (cmsFloat64Number) (j * 65535.0) / (InCurve ->nEntries - 1);
1084 y2 = (cmsFloat64Number) ((j+1) * 65535.0 ) / (InCurve ->nEntries - 1);
1085
1086 // If collapsed, then use any
1087 if (x1 == x2) {
1088
1089 out ->Table16[i] = _cmsQuickSaturateWord(Ascending ? y2 : y1);
1090 continue;
1091
1092 } else {
1093
1094 // Interpolate
1095 a = (y2 - y1) / (x2 - x1);
1096 b = y2 - a * x2;
1097 }
1098 }
1099
1100 out ->Table16[i] = _cmsQuickSaturateWord(a* y + b);
1101 }
1102
1103
1104 return out;
1105 }
1106
1107 // Reverse a gamma table
1108 cmsToneCurve* CMSEXPORT cmsReverseToneCurve(const cmsToneCurve* InGamma)
1109 {
1110 _cmsAssert(InGamma != NULL);
1111
1112 return cmsReverseToneCurveEx(4096, InGamma);
1113 }
1114
1115 // From: Eilers, P.H.C. (1994) Smoothing and interpolation with finite
1116 // differences. in: Graphic Gems IV, Heckbert, P.S. (ed.), Academic press.
1117 //
1118 // Smoothing and interpolation with second differences.
1119 //
1120 // Input: weights (w), data (y): vector from 1 to m.
1121 // Input: smoothing parameter (lambda), length (m).
1122 // Output: smoothed vector (z): vector from 1 to m.
1123
1124 static
1125 cmsBool smooth2(cmsContext ContextID, cmsFloat32Number w[], cmsFloat32Number y[],
1126 cmsFloat32Number z[], cmsFloat32Number lambda, int m)
1127 {
1128 int i, i1, i2;
1129 cmsFloat32Number *c, *d, *e;
1130 cmsBool st;
1131
1132
1133 c = (cmsFloat32Number*) _cmsCalloc(ContextID, MAX_NODES_IN_CURVE, sizeof(cmsFloat32Number));
1134 d = (cmsFloat32Number*) _cmsCalloc(ContextID, MAX_NODES_IN_CURVE, sizeof(cmsFloat32Number));
1135 e = (cmsFloat32Number*) _cmsCalloc(ContextID, MAX_NODES_IN_CURVE, sizeof(cmsFloat32Number));
1136
1137 if (c != NULL && d != NULL && e != NULL) {
1138
1139
1140 d[1] = w[1] + lambda;
1141 c[1] = -2 * lambda / d[1];
1142 e[1] = lambda /d[1];
1143 z[1] = w[1] * y[1];
1144 d[2] = w[2] + 5 * lambda - d[1] * c[1] * c[1];
1145 c[2] = (-4 * lambda - d[1] * c[1] * e[1]) / d[2];
1146 e[2] = lambda / d[2];
1147 z[2] = w[2] * y[2] - c[1] * z[1];
1148
1149 for (i = 3; i < m - 1; i++) {
1150 i1 = i - 1; i2 = i - 2;
1151 d[i]= w[i] + 6 * lambda - c[i1] * c[i1] * d[i1] - e[i2] * e[i2] * d[i2];
1152 c[i] = (-4 * lambda -d[i1] * c[i1] * e[i1])/ d[i];
1153 e[i] = lambda / d[i];
1154 z[i] = w[i] * y[i] - c[i1] * z[i1] - e[i2] * z[i2];
1155 }
1156
1157 i1 = m - 2; i2 = m - 3;
1158
1159 d[m - 1] = w[m - 1] + 5 * lambda -c[i1] * c[i1] * d[i1] - e[i2] * e[i2] * d[i2];
1160 c[m - 1] = (-2 * lambda - d[i1] * c[i1] * e[i1]) / d[m - 1];
1161 z[m - 1] = w[m - 1] * y[m - 1] - c[i1] * z[i1] - e[i2] * z[i2];
1162 i1 = m - 1; i2 = m - 2;
1163
1164 d[m] = w[m] + lambda - c[i1] * c[i1] * d[i1] - e[i2] * e[i2] * d[i2];
1165 z[m] = (w[m] * y[m] - c[i1] * z[i1] - e[i2] * z[i2]) / d[m];
1166 z[m - 1] = z[m - 1] / d[m - 1] - c[m - 1] * z[m];
1167
1168 for (i = m - 2; 1<= i; i--)
1169 z[i] = z[i] / d[i] - c[i] * z[i + 1] - e[i] * z[i + 2];
1170
1171 st = TRUE;
1172 }
1173 else st = FALSE;
1174
1175 if (c != NULL) _cmsFree(ContextID, c);
1176 if (d != NULL) _cmsFree(ContextID, d);
1177 if (e != NULL) _cmsFree(ContextID, e);
1178
1179 return st;
1180 }
1181
1182 // Smooths a curve sampled at regular intervals.
1183 cmsBool CMSEXPORT cmsSmoothToneCurve(cmsToneCurve* Tab, cmsFloat64Number lambda)
1184 {
1185 cmsBool SuccessStatus = TRUE;
1186 cmsFloat32Number *w, *y, *z;
1187 cmsUInt32Number i, nItems, Zeros, Poles;
1188
1189 if (Tab != NULL && Tab->InterpParams != NULL)
1190 {
1191 cmsContext ContextID = Tab->InterpParams->ContextID;
1192
1193 if (!cmsIsToneCurveLinear(Tab)) // Only non-linear curves need smoothing
1194 {
1195 nItems = Tab->nEntries;
1196 if (nItems < MAX_NODES_IN_CURVE)
1197 {
1198 // Allocate one more item than needed
1199 w = (cmsFloat32Number *)_cmsCalloc(ContextID, nItems + 1, sizeof(cmsFloat32Number));
1200 y = (cmsFloat32Number *)_cmsCalloc(ContextID, nItems + 1, sizeof(cmsFloat32Number));
1201 z = (cmsFloat32Number *)_cmsCalloc(ContextID, nItems + 1, sizeof(cmsFloat32Number));
1202
1203 if (w != NULL && y != NULL && z != NULL) // Ensure no memory allocation failure
1204 {
1205 memset(w, 0, (nItems + 1) * sizeof(cmsFloat32Number));
1206 memset(y, 0, (nItems + 1) * sizeof(cmsFloat32Number));
1207 memset(z, 0, (nItems + 1) * sizeof(cmsFloat32Number));
1208
1209 for (i = 0; i < nItems; i++)
1210 {
1211 y[i + 1] = (cmsFloat32Number)Tab->Table16[i];
1212 w[i + 1] = 1.0;
1213 }
1214
1215 if (smooth2(ContextID, w, y, z, (cmsFloat32Number)lambda, (int)nItems))
1216 {
1217 // Do some reality - checking...
1218
1219 Zeros = Poles = 0;
1220 for (i = nItems; i > 1; --i)
1221 {
1222 if (z[i] == 0.) Zeros++;
1223 if (z[i] >= 65535.) Poles++;
1224 if (z[i] < z[i - 1])
1225 {
1226 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Non-Monotonic.");
1227 SuccessStatus = FALSE;
1228 break;
1229 }
1230 }
1231
1232 if (SuccessStatus && Zeros > (nItems / 3))
1233 {
1234 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Degenerated, mostly zeros.");
1235 SuccessStatus = FALSE;
1236 }
1237
1238 if (SuccessStatus && Poles > (nItems / 3))
1239 {
1240 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Degenerated, mostly poles.");
1241 SuccessStatus = FALSE;
1242 }
1243
1244 if (SuccessStatus) // Seems ok
1245 {
1246 for (i = 0; i < nItems; i++)
1247 {
1248 // Clamp to cmsUInt16Number
1249 Tab->Table16[i] = _cmsQuickSaturateWord(z[i + 1]);
1250 }
1251 }
1252 }
1253 else // Could not smooth
1254 {
1255 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Function smooth2 failed.");
1256 SuccessStatus = FALSE;
1257 }
1258 }
1259 else // One or more buffers could not be allocated
1260 {
1261 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Could not allocate memory.");
1262 SuccessStatus = FALSE;
1263 }
1264
1265 if (z != NULL)
1266 _cmsFree(ContextID, z);
1267
1268 if (y != NULL)
1269 _cmsFree(ContextID, y);
1270
1271 if (w != NULL)
1272 _cmsFree(ContextID, w);
1273 }
1274 else // too many items in the table
1275 {
1276 cmsSignalError(ContextID, cmsERROR_RANGE, "cmsSmoothToneCurve: Too many points.");
1277 SuccessStatus = FALSE;
1278 }
1279 }
1280 }
1281 else // Tab parameter or Tab->InterpParams is NULL
1282 {
1283 // Can't signal an error here since the ContextID is not known at this point
1284 SuccessStatus = FALSE;
1285 }
1286
1287 return SuccessStatus;
1288 }
1289
1290 // Is a table linear? Do not use parametric since we cannot guarantee some weird parameters resulting
1291 // in a linear table. This way assures it is linear in 12 bits, which should be enough in most cases.
1292 cmsBool CMSEXPORT cmsIsToneCurveLinear(const cmsToneCurve* Curve)
1293 {
1294 int i;
1295 int diff;
1296
1297 _cmsAssert(Curve != NULL);
1298
1299 for (i=0; i < (int) Curve ->nEntries; i++) {
1300
1301 diff = abs((int) Curve->Table16[i] - (int) _cmsQuantizeVal(i, Curve ->nEntries));
1302 if (diff > 0x0f)
1303 return FALSE;
1304 }
1305
1306 return TRUE;
1307 }
1308
1309 // Same, but for monotonicity
1310 cmsBool CMSEXPORT cmsIsToneCurveMonotonic(const cmsToneCurve* t)
1311 {
1312 cmsUInt32Number n;
1313 int i, last;
1314 cmsBool lDescending;
1315
1316 _cmsAssert(t != NULL);
1317
1318 // Degenerated curves are monotonic? Ok, let's pass them
1319 n = t ->nEntries;
1320 if (n < 2) return TRUE;
1321
1322 // Curve direction
1323 lDescending = cmsIsToneCurveDescending(t);
1324
1325 if (lDescending) {
1326
1327 last = t ->Table16[0];
1328
1329 for (i = 1; i < (int) n; i++) {
1330
1331 if (t ->Table16[i] - last > 2) // We allow some ripple
1332 return FALSE;
1333 else
1334 last = t ->Table16[i];
1335
1336 }
1337 }
1338 else {
1339
1340 last = t ->Table16[n-1];
1341
1342 for (i = (int) n - 2; i >= 0; --i) {
1343
1344 if (t ->Table16[i] - last > 2)
1345 return FALSE;
1346 else
1347 last = t ->Table16[i];
1348
1349 }
1350 }
1351
1352 return TRUE;
1353 }
1354
1355 // Same, but for descending tables
1356 cmsBool CMSEXPORT cmsIsToneCurveDescending(const cmsToneCurve* t)
1357 {
1358 _cmsAssert(t != NULL);
1359
1360 return t ->Table16[0] > t ->Table16[t ->nEntries-1];
1361 }
1362
1363
1364 // Another info fn: is out gamma table multisegment?
1365 cmsBool CMSEXPORT cmsIsToneCurveMultisegment(const cmsToneCurve* t)
1366 {
1367 _cmsAssert(t != NULL);
1368
1369 return t -> nSegments > 1;
1370 }
1371
1372 cmsInt32Number CMSEXPORT cmsGetToneCurveParametricType(const cmsToneCurve* t)
1373 {
1374 _cmsAssert(t != NULL);
1375
1376 if (t -> nSegments != 1) return 0;
1377 return t ->Segments[0].Type;
1378 }
1379
1380 // We need accuracy this time
1381 cmsFloat32Number CMSEXPORT cmsEvalToneCurveFloat(const cmsToneCurve* Curve, cmsFloat32Number v)
1382 {
1383 _cmsAssert(Curve != NULL);
1384
1385 // Check for 16 bits table. If so, this is a limited-precision tone curve
1386 if (Curve ->nSegments == 0) {
1387
1388 cmsUInt16Number In, Out;
1389
1390 In = (cmsUInt16Number) _cmsQuickSaturateWord(v * 65535.0);
1391 Out = cmsEvalToneCurve16(Curve, In);
1392
1393 return (cmsFloat32Number) (Out / 65535.0);
1394 }
1395
1396 return (cmsFloat32Number) EvalSegmentedFn(Curve, v);
1397 }
1398
1399 // We need xput over here
1400 cmsUInt16Number CMSEXPORT cmsEvalToneCurve16(const cmsToneCurve* Curve, cmsUInt16Number v)
1401 {
1402 cmsUInt16Number out;
1403
1404 _cmsAssert(Curve != NULL);
1405
1406 Curve ->InterpParams ->Interpolation.Lerp16(&v, &out, Curve ->InterpParams);
1407 return out;
1408 }
1409
1410
1411 // Least squares fitting.
1412 // A mathematical procedure for finding the best-fitting curve to a given set of points by
1413 // minimizing the sum of the squares of the offsets ("the residuals") of the points from the curve.
1414 // The sum of the squares of the offsets is used instead of the offset absolute values because
1415 // this allows the residuals to be treated as a continuous differentiable quantity.
1416 //
1417 // y = f(x) = x ^ g
1418 //
1419 // R = (yi - (xi^g))
1420 // R2 = (yi - (xi^g))2
1421 // SUM R2 = SUM (yi - (xi^g))2
1422 //
1423 // dR2/dg = -2 SUM x^g log(x)(y - x^g)
1424 // solving for dR2/dg = 0
1425 //
1426 // g = 1/n * SUM(log(y) / log(x))
1427
1428 cmsFloat64Number CMSEXPORT cmsEstimateGamma(const cmsToneCurve* t, cmsFloat64Number Precision)
1429 {
1430 cmsFloat64Number gamma, sum, sum2;
1431 cmsFloat64Number n, x, y, Std;
1432 cmsUInt32Number i;
1433
1434 _cmsAssert(t != NULL);
1435
1436 sum = sum2 = n = 0;
1437
1438 // Excluding endpoints
1439 for (i=1; i < (MAX_NODES_IN_CURVE-1); i++) {
1440
1441 x = (cmsFloat64Number) i / (MAX_NODES_IN_CURVE-1);
1442 y = (cmsFloat64Number) cmsEvalToneCurveFloat(t, (cmsFloat32Number) x);
1443
1444 // Avoid 7% on lower part to prevent
1445 // artifacts due to linear ramps
1446
1447 if (y > 0. && y < 1. && x > 0.07) {
1448
1449 gamma = log(y) / log(x);
1450 sum += gamma;
1451 sum2 += gamma * gamma;
1452 n++;
1453 }
1454 }
1455
1456 // Take a look on SD to see if gamma isn't exponential at all
1457 Std = sqrt((n * sum2 - sum * sum) / (n*(n-1)));
1458
1459 if (Std > Precision)
1460 return -1.0;
1461
1462 return (sum / n); // The mean
1463 }
1464
1465
1466 // Retrieve parameters on one-segment tone curves
1467
1468 cmsFloat64Number* CMSEXPORT cmsGetToneCurveParams(const cmsToneCurve* t)
1469 {
1470 _cmsAssert(t != NULL);
1471
1472 if (t->nSegments != 1) return NULL;
1473 return t->Segments[0].Params;
1474 }