2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6 % DDDD IIIII SSSSS TTTTT OOO RRRR TTTTT %
7 % D D I SS T O O R R T %
8 % D D I SSS T O O RRRR T %
9 % D D I SS T O O R R T %
10 % DDDD IIIII SSSSS T OOO R R T %
13 % MagickCore Image Distortion Methods %
21 % Copyright 1999-2011 ImageMagick Studio LLC, a non-profit organization %
22 % dedicated to making software imaging solutions freely available. %
24 % You may not use this file except in compliance with the License. You may %
25 % obtain a copy of the License at %
27 % http://www.imagemagick.org/script/license.php %
29 % Unless required by applicable law or agreed to in writing, software %
30 % distributed under the License is distributed on an "AS IS" BASIS, %
31 % WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. %
32 % See the License for the specific language governing permissions and %
33 % limitations under the License. %
35 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
43 #include "MagickCore/studio.h"
44 #include "MagickCore/artifact.h"
45 #include "MagickCore/cache.h"
46 #include "MagickCore/cache-view.h"
47 #include "MagickCore/colorspace-private.h"
48 #include "MagickCore/composite-private.h"
49 #include "MagickCore/distort.h"
50 #include "MagickCore/exception.h"
51 #include "MagickCore/exception-private.h"
52 #include "MagickCore/gem.h"
53 #include "MagickCore/hashmap.h"
54 #include "MagickCore/image.h"
55 #include "MagickCore/list.h"
56 #include "MagickCore/matrix.h"
57 #include "MagickCore/memory_.h"
58 #include "MagickCore/monitor-private.h"
59 #include "MagickCore/option.h"
60 #include "MagickCore/pixel.h"
61 #include "MagickCore/pixel-accessor.h"
62 #include "MagickCore/resample.h"
63 #include "MagickCore/resample-private.h"
64 #include "MagickCore/registry.h"
65 #include "MagickCore/semaphore.h"
66 #include "MagickCore/string_.h"
67 #include "MagickCore/string-private.h"
68 #include "MagickCore/thread-private.h"
69 #include "MagickCore/token.h"
70 #include "MagickCore/transform.h"
73 Numerous internal routines for image distortions.
75 static inline double MagickMin(const double x,const double y)
77 return( x < y ? x : y);
79 static inline double MagickMax(const double x,const double y)
81 return( x > y ? x : y);
84 static inline void AffineArgsToCoefficients(double *affine)
86 /* map external sx,ry,rx,sy,tx,ty to internal c0,c2,c4,c1,c3,c5 */
87 double tmp[4]; /* note indexes 0 and 5 remain unchanged */
88 tmp[0]=affine[1]; tmp[1]=affine[2]; tmp[2]=affine[3]; tmp[3]=affine[4];
89 affine[3]=tmp[0]; affine[1]=tmp[1]; affine[4]=tmp[2]; affine[2]=tmp[3];
92 static inline void CoefficientsToAffineArgs(double *coeff)
94 /* map internal c0,c1,c2,c3,c4,c5 to external sx,ry,rx,sy,tx,ty */
95 double tmp[4]; /* note indexes 0 and 5 remain unchanged */
96 tmp[0]=coeff[3]; tmp[1]=coeff[1]; tmp[2]=coeff[4]; tmp[3]=coeff[2];
97 coeff[1]=tmp[0]; coeff[2]=tmp[1]; coeff[3]=tmp[2]; coeff[4]=tmp[3];
99 static void InvertAffineCoefficients(const double *coeff,double *inverse)
101 /* From "Digital Image Warping" by George Wolberg, page 50 */
104 determinant=1.0/(coeff[0]*coeff[4]-coeff[1]*coeff[3]);
105 inverse[0]=determinant*coeff[4];
106 inverse[1]=determinant*(-coeff[1]);
107 inverse[2]=determinant*(coeff[1]*coeff[5]-coeff[2]*coeff[4]);
108 inverse[3]=determinant*(-coeff[3]);
109 inverse[4]=determinant*coeff[0];
110 inverse[5]=determinant*(coeff[2]*coeff[3]-coeff[0]*coeff[5]);
113 static void InvertPerspectiveCoefficients(const double *coeff,
116 /* From "Digital Image Warping" by George Wolberg, page 53 */
119 determinant=1.0/(coeff[0]*coeff[4]-coeff[3]*coeff[1]);
120 inverse[0]=determinant*(coeff[4]-coeff[7]*coeff[5]);
121 inverse[1]=determinant*(coeff[7]*coeff[2]-coeff[1]);
122 inverse[2]=determinant*(coeff[1]*coeff[5]-coeff[4]*coeff[2]);
123 inverse[3]=determinant*(coeff[6]*coeff[5]-coeff[3]);
124 inverse[4]=determinant*(coeff[0]-coeff[6]*coeff[2]);
125 inverse[5]=determinant*(coeff[3]*coeff[2]-coeff[0]*coeff[5]);
126 inverse[6]=determinant*(coeff[3]*coeff[7]-coeff[6]*coeff[4]);
127 inverse[7]=determinant*(coeff[6]*coeff[1]-coeff[0]*coeff[7]);
130 static inline double MagickRound(double x)
133 Round the fraction to nearest integer.
136 return((double) ((ssize_t) (x+0.5)));
137 return((double) ((ssize_t) (x-0.5)));
141 * Polynomial Term Defining Functions
143 * Order must either be an integer, or 1.5 to produce
144 * the 2 number_valuesal polynomial function...
145 * affine 1 (3) u = c0 + c1*x + c2*y
146 * bilinear 1.5 (4) u = '' + c3*x*y
147 * quadratic 2 (6) u = '' + c4*x*x + c5*y*y
148 * cubic 3 (10) u = '' + c6*x^3 + c7*x*x*y + c8*x*y*y + c9*y^3
149 * quartic 4 (15) u = '' + c10*x^4 + ... + c14*y^4
150 * quintic 5 (21) u = '' + c15*x^5 + ... + c20*y^5
151 * number in parenthesis minimum number of points needed.
152 * Anything beyond quintic, has not been implemented until
153 * a more automated way of determining terms is found.
155 * Note the slight re-ordering of the terms for a quadratic polynomial
156 * which is to allow the use of a bi-linear (order=1.5) polynomial.
157 * All the later polynomials are ordered simply from x^N to y^N
159 static size_t poly_number_terms(double order)
161 /* Return the number of terms for a 2d polynomial */
162 if ( order < 1 || order > 5 ||
163 ( order != floor(order) && (order-1.5) > MagickEpsilon) )
164 return 0; /* invalid polynomial order */
165 return((size_t) floor((order+1)*(order+2)/2));
168 static double poly_basis_fn(ssize_t n, double x, double y)
170 /* Return the result for this polynomial term */
172 case 0: return( 1.0 ); /* constant */
174 case 2: return( y ); /* affine order = 1 terms = 3 */
175 case 3: return( x*y ); /* bilinear order = 1.5 terms = 4 */
176 case 4: return( x*x );
177 case 5: return( y*y ); /* quadratic order = 2 terms = 6 */
178 case 6: return( x*x*x );
179 case 7: return( x*x*y );
180 case 8: return( x*y*y );
181 case 9: return( y*y*y ); /* cubic order = 3 terms = 10 */
182 case 10: return( x*x*x*x );
183 case 11: return( x*x*x*y );
184 case 12: return( x*x*y*y );
185 case 13: return( x*y*y*y );
186 case 14: return( y*y*y*y ); /* quartic order = 4 terms = 15 */
187 case 15: return( x*x*x*x*x );
188 case 16: return( x*x*x*x*y );
189 case 17: return( x*x*x*y*y );
190 case 18: return( x*x*y*y*y );
191 case 19: return( x*y*y*y*y );
192 case 20: return( y*y*y*y*y ); /* quintic order = 5 terms = 21 */
194 return( 0 ); /* should never happen */
196 static const char *poly_basis_str(ssize_t n)
198 /* return the result for this polynomial term */
200 case 0: return(""); /* constant */
201 case 1: return("*ii");
202 case 2: return("*jj"); /* affine order = 1 terms = 3 */
203 case 3: return("*ii*jj"); /* bilinear order = 1.5 terms = 4 */
204 case 4: return("*ii*ii");
205 case 5: return("*jj*jj"); /* quadratic order = 2 terms = 6 */
206 case 6: return("*ii*ii*ii");
207 case 7: return("*ii*ii*jj");
208 case 8: return("*ii*jj*jj");
209 case 9: return("*jj*jj*jj"); /* cubic order = 3 terms = 10 */
210 case 10: return("*ii*ii*ii*ii");
211 case 11: return("*ii*ii*ii*jj");
212 case 12: return("*ii*ii*jj*jj");
213 case 13: return("*ii*jj*jj*jj");
214 case 14: return("*jj*jj*jj*jj"); /* quartic order = 4 terms = 15 */
215 case 15: return("*ii*ii*ii*ii*ii");
216 case 16: return("*ii*ii*ii*ii*jj");
217 case 17: return("*ii*ii*ii*jj*jj");
218 case 18: return("*ii*ii*jj*jj*jj");
219 case 19: return("*ii*jj*jj*jj*jj");
220 case 20: return("*jj*jj*jj*jj*jj"); /* quintic order = 5 terms = 21 */
222 return( "UNKNOWN" ); /* should never happen */
224 static double poly_basis_dx(ssize_t n, double x, double y)
226 /* polynomial term for x derivative */
228 case 0: return( 0.0 ); /* constant */
229 case 1: return( 1.0 );
230 case 2: return( 0.0 ); /* affine order = 1 terms = 3 */
231 case 3: return( y ); /* bilinear order = 1.5 terms = 4 */
233 case 5: return( 0.0 ); /* quadratic order = 2 terms = 6 */
234 case 6: return( x*x );
235 case 7: return( x*y );
236 case 8: return( y*y );
237 case 9: return( 0.0 ); /* cubic order = 3 terms = 10 */
238 case 10: return( x*x*x );
239 case 11: return( x*x*y );
240 case 12: return( x*y*y );
241 case 13: return( y*y*y );
242 case 14: return( 0.0 ); /* quartic order = 4 terms = 15 */
243 case 15: return( x*x*x*x );
244 case 16: return( x*x*x*y );
245 case 17: return( x*x*y*y );
246 case 18: return( x*y*y*y );
247 case 19: return( y*y*y*y );
248 case 20: return( 0.0 ); /* quintic order = 5 terms = 21 */
250 return( 0.0 ); /* should never happen */
252 static double poly_basis_dy(ssize_t n, double x, double y)
254 /* polynomial term for y derivative */
256 case 0: return( 0.0 ); /* constant */
257 case 1: return( 0.0 );
258 case 2: return( 1.0 ); /* affine order = 1 terms = 3 */
259 case 3: return( x ); /* bilinear order = 1.5 terms = 4 */
260 case 4: return( 0.0 );
261 case 5: return( y ); /* quadratic order = 2 terms = 6 */
262 default: return( poly_basis_dx(n-1,x,y) ); /* weird but true */
264 /* NOTE: the only reason that last is not true for 'quadratic'
265 is due to the re-arrangement of terms to allow for 'bilinear'
270 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
274 + G e n e r a t e C o e f f i c i e n t s %
278 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
280 % GenerateCoefficients() takes user provided input arguments and generates
281 % the coefficients, needed to apply the specific distortion for either
282 % distorting images (generally using control points) or generating a color
283 % gradient from sparsely separated color points.
285 % The format of the GenerateCoefficients() method is:
287 % Image *GenerateCoefficients(const Image *image,DistortImageMethod method,
288 % const size_t number_arguments,const double *arguments,
289 % size_t number_values, ExceptionInfo *exception)
291 % A description of each parameter follows:
293 % o image: the image to be distorted.
295 % o method: the method of image distortion/ sparse gradient
297 % o number_arguments: the number of arguments given.
299 % o arguments: the arguments for this distortion method.
301 % o number_values: the style and format of given control points, (caller type)
302 % 0: 2 dimensional mapping of control points (Distort)
303 % Format: u,v,x,y where u,v is the 'source' of the
304 % the color to be plotted, for DistortImage()
305 % N: Interpolation of control points with N values (usally r,g,b)
306 % Format: x,y,r,g,b mapping x,y to color values r,g,b
307 % IN future, variable number of values may be given (1 to N)
309 % o exception: return any errors or warnings in this structure
311 % Note that the returned array of double values must be freed by the
312 % calling method using RelinquishMagickMemory(). This however may change in
313 % the future to require a more 'method' specific method.
315 % Because of this this method should not be classed as stable or used
316 % outside other MagickCore library methods.
319 static double *GenerateCoefficients(const Image *image,
320 DistortImageMethod *method,const size_t number_arguments,
321 const double *arguments,size_t number_values,ExceptionInfo *exception)
330 number_coeff, /* number of coefficients to return (array size) */
331 cp_size, /* number floating point numbers per control point */
332 cp_x,cp_y, /* the x,y indexes for control point */
333 cp_values; /* index of values for this control point */
334 /* number_values Number of values given per control point */
336 if ( number_values == 0 ) {
337 /* Image distortion using control points (or other distortion)
338 That is generate a mapping so that x,y->u,v given u,v,x,y
340 number_values = 2; /* special case: two values of u,v */
341 cp_values = 0; /* the values i,j are BEFORE the destination CP x,y */
342 cp_x = 2; /* location of x,y in input control values */
344 /* NOTE: cp_values, also used for later 'reverse map distort' tests */
347 cp_x = 0; /* location of x,y in input control values */
349 cp_values = 2; /* and the other values are after x,y */
350 /* Typically in this case the values are R,G,B color values */
352 cp_size = number_values+2; /* each CP defintion involves this many numbers */
354 /* If not enough control point pairs are found for specific distortions
355 fall back to Affine distortion (allowing 0 to 3 point pairs)
357 if ( number_arguments < 4*cp_size &&
358 ( *method == BilinearForwardDistortion
359 || *method == BilinearReverseDistortion
360 || *method == PerspectiveDistortion
362 *method = AffineDistortion;
366 case AffineDistortion:
367 /* also BarycentricColorInterpolate: */
368 number_coeff=3*number_values;
370 case PolynomialDistortion:
371 /* number of coefficents depend on the given polynomal 'order' */
372 if ( number_arguments <= 1 && (number_arguments-1)%cp_size != 0)
374 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
375 "InvalidArgument","%s : '%s'","Polynomial",
376 "Invalid number of args: order [CPs]...");
377 return((double *) NULL);
379 i = poly_number_terms(arguments[0]);
380 number_coeff = 2 + i*number_values;
382 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
383 "InvalidArgument","%s : '%s'","Polynomial",
384 "Invalid order, should be interger 1 to 5, or 1.5");
385 return((double *) NULL);
387 if ( number_arguments < 1+i*cp_size ) {
388 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
389 "InvalidArgument", "%s : 'require at least %.20g CPs'",
390 "Polynomial", (double) i);
391 return((double *) NULL);
394 case BilinearReverseDistortion:
395 number_coeff=4*number_values;
398 The rest are constants as they are only used for image distorts
400 case BilinearForwardDistortion:
401 number_coeff=10; /* 2*4 coeff plus 2 constants */
402 cp_x = 0; /* Reverse src/dest coords for forward mapping */
407 case QuadraterialDistortion:
408 number_coeff=19; /* BilinearForward + BilinearReverse */
411 case ShepardsDistortion:
412 number_coeff=1; /* not used, but provide some type of return */
417 case ScaleRotateTranslateDistortion:
418 case AffineProjectionDistortion:
419 case Plane2CylinderDistortion:
420 case Cylinder2PlaneDistortion:
423 case PolarDistortion:
424 case DePolarDistortion:
427 case PerspectiveDistortion:
428 case PerspectiveProjectionDistortion:
431 case BarrelDistortion:
432 case BarrelInverseDistortion:
436 assert(! "Unknown Method Given"); /* just fail assertion */
439 /* allocate the array of coefficients needed */
440 coeff = (double *) AcquireQuantumMemory(number_coeff,sizeof(*coeff));
441 if (coeff == (double *) NULL) {
442 (void) ThrowMagickException(exception,GetMagickModule(),
443 ResourceLimitError,"MemoryAllocationFailed",
444 "%s", "GenerateCoefficients");
445 return((double *) NULL);
448 /* zero out coefficients array */
449 for (i=0; i < number_coeff; i++)
454 case AffineDistortion:
458 for each 'value' given
460 Input Arguments are sets of control points...
461 For Distort Images u,v, x,y ...
462 For Sparse Gradients x,y, r,g,b ...
464 if ( number_arguments%cp_size != 0 ||
465 number_arguments < cp_size ) {
466 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
467 "InvalidArgument", "%s : 'require at least %.20g CPs'",
469 coeff=(double *) RelinquishMagickMemory(coeff);
470 return((double *) NULL);
472 /* handle special cases of not enough arguments */
473 if ( number_arguments == cp_size ) {
474 /* Only 1 CP Set Given */
475 if ( cp_values == 0 ) {
476 /* image distortion - translate the image */
478 coeff[2] = arguments[0] - arguments[2];
480 coeff[5] = arguments[1] - arguments[3];
483 /* sparse gradient - use the values directly */
484 for (i=0; i<number_values; i++)
485 coeff[i*3+2] = arguments[cp_values+i];
489 /* 2 or more points (usally 3) given.
490 Solve a least squares simultaneous equation for coefficients.
500 /* create matrix, and a fake vectors matrix */
501 matrix = AcquireMagickMatrix(3UL,3UL);
502 vectors = (double **) AcquireQuantumMemory(number_values,sizeof(*vectors));
503 if (matrix == (double **) NULL || vectors == (double **) NULL)
505 matrix = RelinquishMagickMatrix(matrix, 3UL);
506 vectors = (double **) RelinquishMagickMemory(vectors);
507 coeff = (double *) RelinquishMagickMemory(coeff);
508 (void) ThrowMagickException(exception,GetMagickModule(),
509 ResourceLimitError,"MemoryAllocationFailed",
510 "%s", "DistortCoefficients");
511 return((double *) NULL);
513 /* fake a number_values x3 vectors matrix from coefficients array */
514 for (i=0; i < number_values; i++)
515 vectors[i] = &(coeff[i*3]);
516 /* Add given control point pairs for least squares solving */
517 for (i=0; i < number_arguments; i+=cp_size) {
518 terms[0] = arguments[i+cp_x]; /* x */
519 terms[1] = arguments[i+cp_y]; /* y */
520 terms[2] = 1; /* 1 */
521 LeastSquaresAddTerms(matrix,vectors,terms,
522 &(arguments[i+cp_values]),3UL,number_values);
524 if ( number_arguments == 2*cp_size ) {
525 /* Only two pairs were given, but we need 3 to solve the affine.
526 Fake extra coordinates by rotating p1 around p0 by 90 degrees.
527 x2 = x0 - (y1-y0) y2 = y0 + (x1-x0)
529 terms[0] = arguments[cp_x]
530 - ( arguments[cp_size+cp_y] - arguments[cp_y] ); /* x2 */
531 terms[1] = arguments[cp_y] +
532 + ( arguments[cp_size+cp_x] - arguments[cp_x] ); /* y2 */
533 terms[2] = 1; /* 1 */
534 if ( cp_values == 0 ) {
535 /* Image Distortion - rotate the u,v coordients too */
538 uv2[0] = arguments[0] - arguments[5] + arguments[1]; /* u2 */
539 uv2[1] = arguments[1] + arguments[4] - arguments[0]; /* v2 */
540 LeastSquaresAddTerms(matrix,vectors,terms,uv2,3UL,2UL);
543 /* Sparse Gradient - use values of p0 for linear gradient */
544 LeastSquaresAddTerms(matrix,vectors,terms,
545 &(arguments[cp_values]),3UL,number_values);
548 /* Solve for LeastSquares Coefficients */
549 status=GaussJordanElimination(matrix,vectors,3UL,number_values);
550 matrix = RelinquishMagickMatrix(matrix, 3UL);
551 vectors = (double **) RelinquishMagickMemory(vectors);
552 if ( status == MagickFalse ) {
553 coeff = (double *) RelinquishMagickMemory(coeff);
554 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
555 "InvalidArgument","%s : 'Unsolvable Matrix'",
556 CommandOptionToMnemonic(MagickDistortOptions, *method) );
557 return((double *) NULL);
562 case AffineProjectionDistortion:
565 Arguments: Affine Matrix (forward mapping)
566 Arguments sx, rx, ry, sy, tx, ty
567 Where u = sx*x + ry*y + tx
570 Returns coefficients (in there inverse form) ordered as...
573 AffineProjection Distortion Notes...
574 + Will only work with a 2 number_values for Image Distortion
575 + Can not be used for generating a sparse gradient (interpolation)
578 if (number_arguments != 6) {
579 coeff = (double *) RelinquishMagickMemory(coeff);
580 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
581 "InvalidArgument","%s : 'Needs 6 coeff values'",
582 CommandOptionToMnemonic(MagickDistortOptions, *method) );
583 return((double *) NULL);
585 /* FUTURE: trap test for sx*sy-rx*ry == 0 (determinant = 0, no inverse) */
586 for(i=0; i<6UL; i++ )
587 inverse[i] = arguments[i];
588 AffineArgsToCoefficients(inverse); /* map into coefficents */
589 InvertAffineCoefficients(inverse, coeff); /* invert */
590 *method = AffineDistortion;
594 case ScaleRotateTranslateDistortion:
596 /* Scale, Rotate and Translate Distortion
597 An alternative Affine Distortion
598 Argument options, by number of arguments given:
599 7: x,y, sx,sy, a, nx,ny
606 Where actions are (in order of application)
607 x,y 'center' of transforms (default = image center)
608 sx,sy scale image by this amount (default = 1)
609 a angle of rotation (argument required)
610 nx,ny move 'center' here (default = x,y or no movement)
611 And convert to affine mapping coefficients
613 ScaleRotateTranslate Distortion Notes...
614 + Does not use a set of CPs in any normal way
615 + Will only work with a 2 number_valuesal Image Distortion
616 + Cannot be used for generating a sparse gradient (interpolation)
622 /* set default center, and default scale */
623 x = nx = (double)(image->columns)/2.0 + (double)image->page.x;
624 y = ny = (double)(image->rows)/2.0 + (double)image->page.y;
626 switch ( number_arguments ) {
628 coeff = (double *) RelinquishMagickMemory(coeff);
629 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
630 "InvalidArgument","%s : 'Needs at least 1 argument'",
631 CommandOptionToMnemonic(MagickDistortOptions, *method) );
632 return((double *) NULL);
637 sx = sy = arguments[0];
641 x = nx = arguments[0];
642 y = ny = arguments[1];
643 switch ( number_arguments ) {
648 sx = sy = arguments[2];
657 sx = sy = arguments[2];
670 coeff = (double *) RelinquishMagickMemory(coeff);
671 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
672 "InvalidArgument","%s : 'Too Many Arguments (7 or less)'",
673 CommandOptionToMnemonic(MagickDistortOptions, *method) );
674 return((double *) NULL);
678 /* Trap if sx or sy == 0 -- image is scaled out of existance! */
679 if ( fabs(sx) < MagickEpsilon || fabs(sy) < MagickEpsilon ) {
680 coeff = (double *) RelinquishMagickMemory(coeff);
681 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
682 "InvalidArgument","%s : 'Zero Scale Given'",
683 CommandOptionToMnemonic(MagickDistortOptions, *method) );
684 return((double *) NULL);
686 /* Save the given arguments as an affine distortion */
687 a=DegreesToRadians(a); cosine=cos(a); sine=sin(a);
689 *method = AffineDistortion;
692 coeff[2]=x-nx*coeff[0]-ny*coeff[1];
695 coeff[5]=y-nx*coeff[3]-ny*coeff[4];
698 case PerspectiveDistortion:
700 Perspective Distortion (a ratio of affine distortions)
702 p(x,y) c0*x + c1*y + c2
703 u = ------ = ------------------
704 r(x,y) c6*x + c7*y + 1
706 q(x,y) c3*x + c4*y + c5
707 v = ------ = ------------------
708 r(x,y) c6*x + c7*y + 1
710 c8 = Sign of 'r', or the denominator affine, for the actual image.
711 This determines what part of the distorted image is 'ground'
712 side of the horizon, the other part is 'sky' or invalid.
713 Valid values are +1.0 or -1.0 only.
715 Input Arguments are sets of control points...
716 For Distort Images u,v, x,y ...
717 For Sparse Gradients x,y, r,g,b ...
719 Perspective Distortion Notes...
720 + Can be thought of as ratio of 3 affine transformations
721 + Not separatable: r() or c6 and c7 are used by both equations
722 + All 8 coefficients must be determined simultaniously
723 + Will only work with a 2 number_valuesal Image Distortion
724 + Can not be used for generating a sparse gradient (interpolation)
725 + It is not linear, but is simple to generate an inverse
726 + All lines within an image remain lines.
727 + but distances between points may vary.
741 if ( number_arguments%cp_size != 0 ||
742 number_arguments < cp_size*4 ) {
743 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
744 "InvalidArgument", "%s : 'require at least %.20g CPs'",
745 CommandOptionToMnemonic(MagickDistortOptions, *method), 4.0);
746 coeff=(double *) RelinquishMagickMemory(coeff);
747 return((double *) NULL);
749 /* fake 1x8 vectors matrix directly using the coefficients array */
750 vectors[0] = &(coeff[0]);
751 /* 8x8 least-squares matrix (zeroed) */
752 matrix = AcquireMagickMatrix(8UL,8UL);
753 if (matrix == (double **) NULL) {
754 (void) ThrowMagickException(exception,GetMagickModule(),
755 ResourceLimitError,"MemoryAllocationFailed",
756 "%s", "DistortCoefficients");
757 return((double *) NULL);
759 /* Add control points for least squares solving */
760 for (i=0; i < number_arguments; i+=4) {
761 terms[0]=arguments[i+cp_x]; /* c0*x */
762 terms[1]=arguments[i+cp_y]; /* c1*y */
763 terms[2]=1.0; /* c2*1 */
767 terms[6]=-terms[0]*arguments[i+cp_u]; /* 1/(c6*x) */
768 terms[7]=-terms[1]*arguments[i+cp_u]; /* 1/(c7*y) */
769 LeastSquaresAddTerms(matrix,vectors,terms,&(arguments[i+cp_u]),
775 terms[3]=arguments[i+cp_x]; /* c3*x */
776 terms[4]=arguments[i+cp_y]; /* c4*y */
777 terms[5]=1.0; /* c5*1 */
778 terms[6]=-terms[3]*arguments[i+cp_v]; /* 1/(c6*x) */
779 terms[7]=-terms[4]*arguments[i+cp_v]; /* 1/(c7*y) */
780 LeastSquaresAddTerms(matrix,vectors,terms,&(arguments[i+cp_v]),
783 /* Solve for LeastSquares Coefficients */
784 status=GaussJordanElimination(matrix,vectors,8UL,1UL);
785 matrix = RelinquishMagickMatrix(matrix, 8UL);
786 if ( status == MagickFalse ) {
787 coeff = (double *) RelinquishMagickMemory(coeff);
788 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
789 "InvalidArgument","%s : 'Unsolvable Matrix'",
790 CommandOptionToMnemonic(MagickDistortOptions, *method) );
791 return((double *) NULL);
794 Calculate 9'th coefficient! The ground-sky determination.
795 What is sign of the 'ground' in r() denominator affine function?
796 Just use any valid image coordinate (first control point) in
797 destination for determination of what part of view is 'ground'.
799 coeff[8] = coeff[6]*arguments[cp_x]
800 + coeff[7]*arguments[cp_y] + 1.0;
801 coeff[8] = (coeff[8] < 0.0) ? -1.0 : +1.0;
805 case PerspectiveProjectionDistortion:
808 Arguments: Perspective Coefficents (forward mapping)
810 if (number_arguments != 8) {
811 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
812 "InvalidArgument", "%s : 'Needs 8 coefficient values'",
813 CommandOptionToMnemonic(MagickDistortOptions, *method));
814 return((double *) NULL);
816 /* FUTURE: trap test c0*c4-c3*c1 == 0 (determinate = 0, no inverse) */
817 InvertPerspectiveCoefficients(arguments, coeff);
819 Calculate 9'th coefficient! The ground-sky determination.
820 What is sign of the 'ground' in r() denominator affine function?
821 Just use any valid image cocodinate in destination for determination.
822 For a forward mapped perspective the images 0,0 coord will map to
823 c2,c5 in the distorted image, so set the sign of denominator of that.
825 coeff[8] = coeff[6]*arguments[2]
826 + coeff[7]*arguments[5] + 1.0;
827 coeff[8] = (coeff[8] < 0.0) ? -1.0 : +1.0;
828 *method = PerspectiveDistortion;
832 case BilinearForwardDistortion:
833 case BilinearReverseDistortion:
835 /* Bilinear Distortion (Forward mapping)
836 v = c0*x + c1*y + c2*x*y + c3;
837 for each 'value' given
839 This is actually a simple polynomial Distortion! The difference
840 however is when we need to reverse the above equation to generate a
841 BilinearForwardDistortion (see below).
843 Input Arguments are sets of control points...
844 For Distort Images u,v, x,y ...
845 For Sparse Gradients x,y, r,g,b ...
856 /* check the number of arguments */
857 if ( number_arguments%cp_size != 0 ||
858 number_arguments < cp_size*4 ) {
859 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
860 "InvalidArgument", "%s : 'require at least %.20g CPs'",
861 CommandOptionToMnemonic(MagickDistortOptions, *method), 4.0);
862 coeff=(double *) RelinquishMagickMemory(coeff);
863 return((double *) NULL);
865 /* create matrix, and a fake vectors matrix */
866 matrix = AcquireMagickMatrix(4UL,4UL);
867 vectors = (double **) AcquireQuantumMemory(number_values,sizeof(*vectors));
868 if (matrix == (double **) NULL || vectors == (double **) NULL)
870 matrix = RelinquishMagickMatrix(matrix, 4UL);
871 vectors = (double **) RelinquishMagickMemory(vectors);
872 coeff = (double *) RelinquishMagickMemory(coeff);
873 (void) ThrowMagickException(exception,GetMagickModule(),
874 ResourceLimitError,"MemoryAllocationFailed",
875 "%s", "DistortCoefficients");
876 return((double *) NULL);
878 /* fake a number_values x4 vectors matrix from coefficients array */
879 for (i=0; i < number_values; i++)
880 vectors[i] = &(coeff[i*4]);
881 /* Add given control point pairs for least squares solving */
882 for (i=0; i < number_arguments; i+=cp_size) {
883 terms[0] = arguments[i+cp_x]; /* x */
884 terms[1] = arguments[i+cp_y]; /* y */
885 terms[2] = terms[0]*terms[1]; /* x*y */
886 terms[3] = 1; /* 1 */
887 LeastSquaresAddTerms(matrix,vectors,terms,
888 &(arguments[i+cp_values]),4UL,number_values);
890 /* Solve for LeastSquares Coefficients */
891 status=GaussJordanElimination(matrix,vectors,4UL,number_values);
892 matrix = RelinquishMagickMatrix(matrix, 4UL);
893 vectors = (double **) RelinquishMagickMemory(vectors);
894 if ( status == MagickFalse ) {
895 coeff = (double *) RelinquishMagickMemory(coeff);
896 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
897 "InvalidArgument","%s : 'Unsolvable Matrix'",
898 CommandOptionToMnemonic(MagickDistortOptions, *method) );
899 return((double *) NULL);
901 if ( *method == BilinearForwardDistortion ) {
902 /* Bilinear Forward Mapped Distortion
904 The above least-squares solved for coefficents but in the forward
905 direction, due to changes to indexing constants.
907 i = c0*x + c1*y + c2*x*y + c3;
908 j = c4*x + c5*y + c6*x*y + c7;
910 where i,j are in the destination image, NOT the source.
912 Reverse Pixel mapping however needs to use reverse of these
913 functions. It required a full page of algbra to work out the
914 reversed mapping formula, but resolves down to the following...
917 c9 = 2*(c2*c5-c1*c6); // '2*a' in the quadratic formula
919 i = i - c3; j = j - c7;
920 b = c6*i - c2*j + c8; // So that a*y^2 + b*y + c == 0
921 c = c4*i - c0*j; // y = ( -b +- sqrt(bb - 4ac) ) / (2*a)
925 y = ( -b + sqrt(r) ) / c9;
929 x = ( i - c1*y) / ( c1 - c2*y );
931 NB: if 'r' is negative there is no solution!
932 NB: the sign of the sqrt() should be negative if image becomes
933 flipped or flopped, or crosses over itself.
934 NB: techniqually coefficient c5 is not needed, anymore,
935 but kept for completness.
937 See Anthony Thyssen <A.Thyssen@griffith.edu.au>
938 or Fred Weinhaus <fmw@alink.net> for more details.
941 coeff[8] = coeff[0]*coeff[5] - coeff[1]*coeff[4];
942 coeff[9] = 2*(coeff[2]*coeff[5] - coeff[1]*coeff[6]);
947 case QuadrilateralDistortion:
949 /* Map a Quadrilateral to a unit square using BilinearReverse
950 Then map that unit square back to the final Quadrilateral
951 using BilinearForward.
953 Input Arguments are sets of control points...
954 For Distort Images u,v, x,y ...
955 For Sparse Gradients x,y, r,g,b ...
958 /* UNDER CONSTRUCTION */
963 case PolynomialDistortion:
965 /* Polynomial Distortion
967 First two coefficents are used to hole global polynomal information
968 c0 = Order of the polynimial being created
969 c1 = number_of_terms in one polynomial equation
971 Rest of the coefficients map to the equations....
972 v = c0 + c1*x + c2*y + c3*x*y + c4*x^2 + c5*y^2 + c6*x^3 + ...
973 for each 'value' (number_values of them) given.
974 As such total coefficients = 2 + number_terms * number_values
976 Input Arguments are sets of control points...
977 For Distort Images order [u,v, x,y] ...
978 For Sparse Gradients order [x,y, r,g,b] ...
980 Polynomial Distortion Notes...
981 + UNDER DEVELOPMENT -- Do not expect this to remain as is.
982 + Currently polynomial is a reversed mapped distortion.
983 + Order 1.5 is fudged to map into a bilinear distortion.
984 though it is not the same order as that distortion.
992 nterms; /* number of polynomial terms per number_values */
1000 /* first two coefficients hold polynomial order information */
1001 coeff[0] = arguments[0];
1002 coeff[1] = (double) poly_number_terms(arguments[0]);
1003 nterms = (size_t) coeff[1];
1005 /* create matrix, a fake vectors matrix, and least sqs terms */
1006 matrix = AcquireMagickMatrix(nterms,nterms);
1007 vectors = (double **) AcquireQuantumMemory(number_values,sizeof(*vectors));
1008 terms = (double *) AcquireQuantumMemory(nterms, sizeof(*terms));
1009 if (matrix == (double **) NULL ||
1010 vectors == (double **) NULL ||
1011 terms == (double *) NULL )
1013 matrix = RelinquishMagickMatrix(matrix, nterms);
1014 vectors = (double **) RelinquishMagickMemory(vectors);
1015 terms = (double *) RelinquishMagickMemory(terms);
1016 coeff = (double *) RelinquishMagickMemory(coeff);
1017 (void) ThrowMagickException(exception,GetMagickModule(),
1018 ResourceLimitError,"MemoryAllocationFailed",
1019 "%s", "DistortCoefficients");
1020 return((double *) NULL);
1022 /* fake a number_values x3 vectors matrix from coefficients array */
1023 for (i=0; i < number_values; i++)
1024 vectors[i] = &(coeff[2+i*nterms]);
1025 /* Add given control point pairs for least squares solving */
1026 for (i=1; i < number_arguments; i+=cp_size) { /* NB: start = 1 not 0 */
1027 for (j=0; j < (ssize_t) nterms; j++)
1028 terms[j] = poly_basis_fn(j,arguments[i+cp_x],arguments[i+cp_y]);
1029 LeastSquaresAddTerms(matrix,vectors,terms,
1030 &(arguments[i+cp_values]),nterms,number_values);
1032 terms = (double *) RelinquishMagickMemory(terms);
1033 /* Solve for LeastSquares Coefficients */
1034 status=GaussJordanElimination(matrix,vectors,nterms,number_values);
1035 matrix = RelinquishMagickMatrix(matrix, nterms);
1036 vectors = (double **) RelinquishMagickMemory(vectors);
1037 if ( status == MagickFalse ) {
1038 coeff = (double *) RelinquishMagickMemory(coeff);
1039 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1040 "InvalidArgument","%s : 'Unsolvable Matrix'",
1041 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1042 return((double *) NULL);
1049 Args: arc_width rotate top_edge_radius bottom_edge_radius
1050 All but first argument are optional
1051 arc_width The angle over which to arc the image side-to-side
1052 rotate Angle to rotate image from vertical center
1053 top_radius Set top edge of source image at this radius
1054 bottom_radius Set bootom edge to this radius (radial scaling)
1056 By default, if the radii arguments are nor provided the image radius
1057 is calculated so the horizontal center-line is fits the given arc
1060 The output image size is ALWAYS adjusted to contain the whole image,
1061 and an offset is given to position image relative to the 0,0 point of
1062 the origin, allowing users to use relative positioning onto larger
1063 background (via -flatten).
1065 The arguments are converted to these coefficients
1066 c0: angle for center of source image
1067 c1: angle scale for mapping to source image
1068 c2: radius for top of source image
1069 c3: radius scale for mapping source image
1070 c4: centerline of arc within source image
1072 Note the coefficients use a center angle, so asymptotic join is
1073 furthest from both sides of the source image. This also means that
1074 for arc angles greater than 360 the sides of the image will be
1077 Arc Distortion Notes...
1078 + Does not use a set of CPs
1079 + Will only work with Image Distortion
1080 + Can not be used for generating a sparse gradient (interpolation)
1082 if ( number_arguments >= 1 && arguments[0] < MagickEpsilon ) {
1083 coeff = (double *) RelinquishMagickMemory(coeff);
1084 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1085 "InvalidArgument","%s : 'Arc Angle Too Small'",
1086 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1087 return((double *) NULL);
1089 if ( number_arguments >= 3 && arguments[2] < MagickEpsilon ) {
1090 coeff = (double *) RelinquishMagickMemory(coeff);
1091 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1092 "InvalidArgument","%s : 'Outer Radius Too Small'",
1093 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1094 return((double *) NULL);
1096 coeff[0] = -MagickPI2; /* -90, place at top! */
1097 if ( number_arguments >= 1 )
1098 coeff[1] = DegreesToRadians(arguments[0]);
1100 coeff[1] = MagickPI2; /* zero arguments - center is at top */
1101 if ( number_arguments >= 2 )
1102 coeff[0] += DegreesToRadians(arguments[1]);
1103 coeff[0] /= Magick2PI; /* normalize radians */
1104 coeff[0] -= MagickRound(coeff[0]);
1105 coeff[0] *= Magick2PI; /* de-normalize back to radians */
1106 coeff[3] = (double)image->rows-1;
1107 coeff[2] = (double)image->columns/coeff[1] + coeff[3]/2.0;
1108 if ( number_arguments >= 3 ) {
1109 if ( number_arguments >= 4 )
1110 coeff[3] = arguments[2] - arguments[3];
1112 coeff[3] *= arguments[2]/coeff[2];
1113 coeff[2] = arguments[2];
1115 coeff[4] = ((double)image->columns-1.0)/2.0;
1119 case PolarDistortion:
1120 case DePolarDistortion:
1122 /* (De)Polar Distortion (same set of arguments)
1123 Args: Rmax, Rmin, Xcenter,Ycenter, Afrom,Ato
1124 DePolar can also have the extra arguments of Width, Height
1126 Coefficients 0 to 5 is the sanatized version first 6 input args
1127 Coefficient 6 is the angle to coord ratio and visa-versa
1128 Coefficient 7 is the radius to coord ratio and visa-versa
1130 WARNING: It is possible for Radius max<min and/or Angle from>to
1132 if ( number_arguments == 3
1133 || ( number_arguments > 6 && *method == PolarDistortion )
1134 || number_arguments > 8 ) {
1135 (void) ThrowMagickException(exception,GetMagickModule(),
1136 OptionError,"InvalidArgument", "%s : number of arguments",
1137 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1138 coeff=(double *) RelinquishMagickMemory(coeff);
1139 return((double *) NULL);
1141 /* Rmax - if 0 calculate appropriate value */
1142 if ( number_arguments >= 1 )
1143 coeff[0] = arguments[0];
1146 /* Rmin - usally 0 */
1147 coeff[1] = number_arguments >= 2 ? arguments[1] : 0.0;
1149 if ( number_arguments >= 4 ) {
1150 coeff[2] = arguments[2];
1151 coeff[3] = arguments[3];
1153 else { /* center of actual image */
1154 coeff[2] = (double)(image->columns)/2.0+image->page.x;
1155 coeff[3] = (double)(image->rows)/2.0+image->page.y;
1157 /* Angle from,to - about polar center 0 is downward */
1158 coeff[4] = -MagickPI;
1159 if ( number_arguments >= 5 )
1160 coeff[4] = DegreesToRadians(arguments[4]);
1161 coeff[5] = coeff[4];
1162 if ( number_arguments >= 6 )
1163 coeff[5] = DegreesToRadians(arguments[5]);
1164 if ( fabs(coeff[4]-coeff[5]) < MagickEpsilon )
1165 coeff[5] += Magick2PI; /* same angle is a full circle */
1166 /* if radius 0 or negative, its a special value... */
1167 if ( coeff[0] < MagickEpsilon ) {
1168 /* Use closest edge if radius == 0 */
1169 if ( fabs(coeff[0]) < MagickEpsilon ) {
1170 coeff[0]=MagickMin(fabs(coeff[2]-image->page.x),
1171 fabs(coeff[3]-image->page.y));
1172 coeff[0]=MagickMin(coeff[0],
1173 fabs(coeff[2]-image->page.x-image->columns));
1174 coeff[0]=MagickMin(coeff[0],
1175 fabs(coeff[3]-image->page.y-image->rows));
1177 /* furthest diagonal if radius == -1 */
1178 if ( fabs(-1.0-coeff[0]) < MagickEpsilon ) {
1180 rx = coeff[2]-image->page.x;
1181 ry = coeff[3]-image->page.y;
1182 coeff[0] = rx*rx+ry*ry;
1183 ry = coeff[3]-image->page.y-image->rows;
1184 coeff[0] = MagickMax(coeff[0],rx*rx+ry*ry);
1185 rx = coeff[2]-image->page.x-image->columns;
1186 coeff[0] = MagickMax(coeff[0],rx*rx+ry*ry);
1187 ry = coeff[3]-image->page.y;
1188 coeff[0] = MagickMax(coeff[0],rx*rx+ry*ry);
1189 coeff[0] = sqrt(coeff[0]);
1192 /* IF Rmax <= 0 or Rmin < 0 OR Rmax < Rmin, THEN error */
1193 if ( coeff[0] < MagickEpsilon || coeff[1] < -MagickEpsilon
1194 || (coeff[0]-coeff[1]) < MagickEpsilon ) {
1195 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1196 "InvalidArgument", "%s : Invalid Radius",
1197 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1198 coeff=(double *) RelinquishMagickMemory(coeff);
1199 return((double *) NULL);
1201 /* converstion ratios */
1202 if ( *method == PolarDistortion ) {
1203 coeff[6]=(double) image->columns/(coeff[5]-coeff[4]);
1204 coeff[7]=(double) image->rows/(coeff[0]-coeff[1]);
1206 else { /* *method == DePolarDistortion */
1207 coeff[6]=(coeff[5]-coeff[4])/image->columns;
1208 coeff[7]=(coeff[0]-coeff[1])/image->rows;
1212 case Cylinder2PlaneDistortion:
1213 case Plane2CylinderDistortion:
1215 /* 3D Cylinder to/from a Tangential Plane
1217 Projection between a clinder and flat plain from a point on the
1218 center line of the cylinder.
1220 The two surfaces coincide in 3D space at the given centers of
1221 distortion (perpendicular to projection point) on both images.
1224 Coefficents: FOV(radians), Radius, center_x,y, dest_center_x,y
1226 FOV (Field Of View) the angular field of view of the distortion,
1227 across the width of the image, in degrees. The centers are the
1228 points of least distortion in the input and resulting images.
1230 These centers are however determined later.
1232 Coeff 0 is the FOV angle of view of image width in radians
1233 Coeff 1 is calculated radius of cylinder.
1234 Coeff 2,3 center of distortion of input image
1235 Coefficents 4,5 Center of Distortion of dest (determined later)
1237 if ( arguments[0] < MagickEpsilon || arguments[0] > 160.0 ) {
1238 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1239 "InvalidArgument", "%s : Invalid FOV Angle",
1240 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1241 coeff=(double *) RelinquishMagickMemory(coeff);
1242 return((double *) NULL);
1244 coeff[0] = DegreesToRadians(arguments[0]);
1245 if ( *method == Cylinder2PlaneDistortion )
1246 /* image is curved around cylinder, so FOV angle (in radians)
1247 * scales directly to image X coordinate, according to its radius.
1249 coeff[1] = (double) image->columns/coeff[0];
1251 /* radius is distance away from an image with this angular FOV */
1252 coeff[1] = (double) image->columns / ( 2 * tan(coeff[0]/2) );
1254 coeff[2] = (double)(image->columns)/2.0+image->page.x;
1255 coeff[3] = (double)(image->rows)/2.0+image->page.y;
1256 coeff[4] = coeff[2];
1257 coeff[5] = coeff[3]; /* assuming image size is the same */
1260 case BarrelDistortion:
1261 case BarrelInverseDistortion:
1263 /* Barrel Distortion
1264 Rs=(A*Rd^3 + B*Rd^2 + C*Rd + D)*Rd
1265 BarrelInv Distortion
1266 Rs=Rd/(A*Rd^3 + B*Rd^2 + C*Rd + D)
1268 Where Rd is the normalized radius from corner to middle of image
1269 Input Arguments are one of the following forms (number of arguments)...
1274 8: Ax,Bx,Cx,Dx Ay,By,Cy,Dy
1275 10: Ax,Bx,Cx,Dx Ay,By,Cy,Dy X,Y
1277 Returns 10 coefficent values, which are de-normalized (pixel scale)
1278 Ax, Bx, Cx, Dx, Ay, By, Cy, Dy, Xc, Yc
1280 /* Radius de-normalization scaling factor */
1282 rscale = 2.0/MagickMin((double) image->columns,(double) image->rows);
1284 /* sanity check number of args must = 3,4,5,6,8,10 or error */
1285 if ( (number_arguments < 3) || (number_arguments == 7) ||
1286 (number_arguments == 9) || (number_arguments > 10) )
1288 coeff=(double *) RelinquishMagickMemory(coeff);
1289 (void) ThrowMagickException(exception,GetMagickModule(),
1290 OptionError,"InvalidArgument", "%s : number of arguments",
1291 CommandOptionToMnemonic(MagickDistortOptions, *method) );
1292 return((double *) NULL);
1294 /* A,B,C,D coefficients */
1295 coeff[0] = arguments[0];
1296 coeff[1] = arguments[1];
1297 coeff[2] = arguments[2];
1298 if ((number_arguments == 3) || (number_arguments == 5) )
1299 coeff[3] = 1.0 - coeff[0] - coeff[1] - coeff[2];
1301 coeff[3] = arguments[3];
1302 /* de-normalize the coefficients */
1303 coeff[0] *= pow(rscale,3.0);
1304 coeff[1] *= rscale*rscale;
1306 /* Y coefficients: as given OR same as X coefficients */
1307 if ( number_arguments >= 8 ) {
1308 coeff[4] = arguments[4] * pow(rscale,3.0);
1309 coeff[5] = arguments[5] * rscale*rscale;
1310 coeff[6] = arguments[6] * rscale;
1311 coeff[7] = arguments[7];
1314 coeff[4] = coeff[0];
1315 coeff[5] = coeff[1];
1316 coeff[6] = coeff[2];
1317 coeff[7] = coeff[3];
1319 /* X,Y Center of Distortion (image coodinates) */
1320 if ( number_arguments == 5 ) {
1321 coeff[8] = arguments[3];
1322 coeff[9] = arguments[4];
1324 else if ( number_arguments == 6 ) {
1325 coeff[8] = arguments[4];
1326 coeff[9] = arguments[5];
1328 else if ( number_arguments == 10 ) {
1329 coeff[8] = arguments[8];
1330 coeff[9] = arguments[9];
1333 /* center of the image provided (image coodinates) */
1334 coeff[8] = (double)image->columns/2.0 + image->page.x;
1335 coeff[9] = (double)image->rows/2.0 + image->page.y;
1339 case ShepardsDistortion:
1341 /* Shepards Distortion input arguments are the coefficents!
1342 Just check the number of arguments is valid!
1343 Args: u1,v1, x1,y1, ...
1344 OR : u1,v1, r1,g1,c1, ...
1346 if ( number_arguments%cp_size != 0 ||
1347 number_arguments < cp_size ) {
1348 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1349 "InvalidArgument", "%s : 'require at least %.20g CPs'",
1350 CommandOptionToMnemonic(MagickDistortOptions, *method), 1.0);
1351 coeff=(double *) RelinquishMagickMemory(coeff);
1352 return((double *) NULL);
1359 /* you should never reach this point */
1360 assert(! "No Method Handler"); /* just fail assertion */
1361 return((double *) NULL);
1365 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1369 + D i s t o r t R e s i z e I m a g e %
1373 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1375 % DistortResizeImage() resize image using the equivalent but slower image
1376 % distortion operator. The filter is applied using a EWA cylindrical
1377 % resampling. But like resize the final image size is limited to whole pixels
1378 % with no effects by virtual-pixels on the result.
1380 % Note that images containing a transparency channel will be twice as slow to
1381 % resize as images one without transparency.
1383 % The format of the DistortResizeImage method is:
1385 % Image *AdaptiveResizeImage(const Image *image,const size_t columns,
1386 % const size_t rows,ExceptionInfo *exception)
1388 % A description of each parameter follows:
1390 % o image: the image.
1392 % o columns: the number of columns in the resized image.
1394 % o rows: the number of rows in the resized image.
1396 % o exception: return any errors or warnings in this structure.
1399 MagickExport Image *DistortResizeImage(const Image *image,
1400 const size_t columns,const size_t rows,ExceptionInfo *exception)
1402 #define DistortResizeImageTag "Distort/Image"
1418 Distort resize image.
1420 assert(image != (const Image *) NULL);
1421 assert(image->signature == MagickSignature);
1422 if (image->debug != MagickFalse)
1423 (void) LogMagickEvent(TraceEvent,GetMagickModule(),"%s",image->filename);
1424 assert(exception != (ExceptionInfo *) NULL);
1425 assert(exception->signature == MagickSignature);
1426 if ((columns == 0) || (rows == 0))
1427 return((Image *) NULL);
1428 /* Do not short-circuit this resize if final image size is unchanged */
1430 (void) SetImageVirtualPixelMethod(image,TransparentVirtualPixelMethod);
1432 (void) ResetMagickMemory(distort_args,0,12*sizeof(double));
1433 distort_args[4]=(double) image->columns;
1434 distort_args[6]=(double) columns;
1435 distort_args[9]=(double) image->rows;
1436 distort_args[11]=(double) rows;
1438 vp_save=GetImageVirtualPixelMethod(image);
1440 tmp_image=CloneImage(image,0,0,MagickTrue,exception);
1441 if ( tmp_image == (Image *) NULL )
1442 return((Image *) NULL);
1443 (void) SetImageVirtualPixelMethod(tmp_image,TransparentVirtualPixelMethod);
1445 if (image->matte == MagickFalse)
1448 Image has not transparency channel, so we free to use it
1450 (void) SetImageAlphaChannel(tmp_image,SetAlphaChannel);
1451 resize_image=DistortImage(tmp_image,AffineDistortion,12,distort_args,
1452 MagickTrue,exception),
1454 tmp_image=DestroyImage(tmp_image);
1455 if ( resize_image == (Image *) NULL )
1456 return((Image *) NULL);
1458 (void) SetImageAlphaChannel(resize_image,DeactivateAlphaChannel);
1459 InheritException(exception,&image->exception);
1464 Image has transparency so handle colors and alpha separatly.
1465 Basically we need to separate Virtual-Pixel alpha in the resized
1466 image, so only the actual original images alpha channel is used.
1471 /* distort alpha channel separately */
1472 (void) SeparateImageChannel(tmp_image,TrueAlphaChannel);
1473 (void) SetImageAlphaChannel(tmp_image,OpaqueAlphaChannel);
1474 resize_alpha=DistortImage(tmp_image,AffineDistortion,12,distort_args,
1475 MagickTrue,exception),
1476 tmp_image=DestroyImage(tmp_image);
1477 if ( resize_alpha == (Image *) NULL )
1478 return((Image *) NULL);
1480 /* distort the actual image containing alpha + VP alpha */
1481 tmp_image=CloneImage(image,0,0,MagickTrue,exception);
1482 if ( tmp_image == (Image *) NULL )
1483 return((Image *) NULL);
1484 (void) SetImageVirtualPixelMethod(tmp_image,
1485 TransparentVirtualPixelMethod);
1486 resize_image=DistortImage(tmp_image,AffineDistortion,12,distort_args,
1487 MagickTrue,exception),
1488 tmp_image=DestroyImage(tmp_image);
1489 if ( resize_image == (Image *) NULL)
1491 resize_alpha=DestroyImage(resize_alpha);
1492 return((Image *) NULL);
1495 /* replace resize images alpha with the separally distorted alpha */
1496 (void) SetImageAlphaChannel(resize_image,DeactivateAlphaChannel);
1497 (void) SetImageAlphaChannel(resize_alpha,DeactivateAlphaChannel);
1498 (void) CompositeImage(resize_image,CopyOpacityCompositeOp,resize_alpha,
1500 InheritException(exception,&resize_image->exception);
1501 resize_alpha=DestroyImage(resize_alpha);
1503 (void) SetImageVirtualPixelMethod(resize_image,vp_save);
1506 Clean up the results of the Distortion
1508 crop_area.width=columns;
1509 crop_area.height=rows;
1513 tmp_image=resize_image;
1514 resize_image=CropImage(tmp_image,&crop_area,exception);
1515 tmp_image=DestroyImage(tmp_image);
1517 if ( resize_image == (Image *) NULL )
1518 return((Image *) NULL);
1520 return(resize_image);
1524 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1528 % D i s t o r t I m a g e %
1532 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1534 % DistortImage() distorts an image using various distortion methods, by
1535 % mapping color lookups of the source image to a new destination image
1536 % usally of the same size as the source image, unless 'bestfit' is set to
1539 % If 'bestfit' is enabled, and distortion allows it, the destination image is
1540 % adjusted to ensure the whole source 'image' will just fit within the final
1541 % destination image, which will be sized and offset accordingly. Also in
1542 % many cases the virtual offset of the source image will be taken into
1543 % account in the mapping.
1545 % If the '-verbose' control option has been set print to standard error the
1546 % equicelent '-fx' formula with coefficients for the function, if practical.
1548 % The format of the DistortImage() method is:
1550 % Image *DistortImage(const Image *image,const DistortImageMethod method,
1551 % const size_t number_arguments,const double *arguments,
1552 % MagickBooleanType bestfit, ExceptionInfo *exception)
1554 % A description of each parameter follows:
1556 % o image: the image to be distorted.
1558 % o method: the method of image distortion.
1560 % ArcDistortion always ignores source image offset, and always
1561 % 'bestfit' the destination image with the top left corner offset
1562 % relative to the polar mapping center.
1564 % Affine, Perspective, and Bilinear, do least squares fitting of the
1565 % distrotion when more than the minimum number of control point pairs
1568 % Perspective, and Bilinear, fall back to a Affine distortion when less
1569 % than 4 control point pairs are provided. While Affine distortions
1570 % let you use any number of control point pairs, that is Zero pairs is
1571 % a No-Op (viewport only) distortion, one pair is a translation and
1572 % two pairs of control points do a scale-rotate-translate, without any
1575 % o number_arguments: the number of arguments given.
1577 % o arguments: an array of floating point arguments for this method.
1579 % o bestfit: Attempt to 'bestfit' the size of the resulting image.
1580 % This also forces the resulting image to be a 'layered' virtual
1581 % canvas image. Can be overridden using 'distort:viewport' setting.
1583 % o exception: return any errors or warnings in this structure
1585 % Extra Controls from Image meta-data (artifacts)...
1588 % Output to stderr alternatives, internal coefficents, and FX
1589 % equivalents for the distortion operation (if feasible).
1590 % This forms an extra check of the distortion method, and allows users
1591 % access to the internal constants IM calculates for the distortion.
1593 % o "distort:viewport"
1594 % Directly set the output image canvas area and offest to use for the
1595 % resulting image, rather than use the original images canvas, or a
1596 % calculated 'bestfit' canvas.
1599 % Scale the size of the output canvas by this amount to provide a
1600 % method of Zooming, and for super-sampling the results.
1602 % Other settings that can effect results include
1604 % o 'interpolate' For source image lookups (scale enlargements)
1606 % o 'filter' Set filter to use for area-resampling (scale shrinking).
1607 % Set to 'point' to turn off and use 'interpolate' lookup
1612 MagickExport Image *DistortImage(const Image *image,DistortImageMethod method,
1613 const size_t number_arguments,const double *arguments,
1614 MagickBooleanType bestfit,ExceptionInfo *exception)
1616 #define DistortImageTag "Distort/Image"
1626 geometry; /* geometry of the distorted space viewport */
1631 assert(image != (Image *) NULL);
1632 assert(image->signature == MagickSignature);
1633 if (image->debug != MagickFalse)
1634 (void) LogMagickEvent(TraceEvent,GetMagickModule(),"%s",image->filename);
1635 assert(exception != (ExceptionInfo *) NULL);
1636 assert(exception->signature == MagickSignature);
1640 Handle Special Compound Distortions
1642 if ( method == ResizeDistortion )
1644 if ( number_arguments != 2 )
1646 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
1647 "InvalidArgument","%s : '%s'","Resize",
1648 "Invalid number of args: 2 only");
1649 return((Image *) NULL);
1651 distort_image=DistortResizeImage(image,(size_t)arguments[0],
1652 (size_t)arguments[1], exception);
1653 return(distort_image);
1657 Convert input arguments (usually as control points for reverse mapping)
1658 into mapping coefficients to apply the distortion.
1660 Note that some distortions are mapped to other distortions,
1661 and as such do not require specific code after this point.
1663 coeff = GenerateCoefficients(image, &method, number_arguments,
1664 arguments, 0, exception);
1665 if ( coeff == (double *) NULL )
1666 return((Image *) NULL);
1669 Determine the size and offset for a 'bestfit' destination.
1670 Usally the four corners of the source image is enough.
1673 /* default output image bounds, when no 'bestfit' is requested */
1674 geometry.width=image->columns;
1675 geometry.height=image->rows;
1679 if ( method == ArcDistortion ) {
1680 bestfit = MagickTrue; /* always calculate a 'best fit' viewport */
1683 /* Work out the 'best fit', (required for ArcDistortion) */
1686 s,d,min,max; /* source, dest coords --mapping--> min, max coords */
1689 fix_bounds = MagickTrue; /* enlarge bounds for VP handling */
1691 s.x=s.y=min.x=max.x=min.y=max.y=0.0; /* keep compiler happy */
1693 /* defines to figure out the bounds of the distorted image */
1694 #define InitalBounds(p) \
1696 /* printf("%lg,%lg -> %lg,%lg\n", s.x,s.y, d.x,d.y); */ \
1697 min.x = max.x = p.x; \
1698 min.y = max.y = p.y; \
1700 #define ExpandBounds(p) \
1702 /* printf("%lg,%lg -> %lg,%lg\n", s.x,s.y, d.x,d.y); */ \
1703 min.x = MagickMin(min.x,p.x); \
1704 max.x = MagickMax(max.x,p.x); \
1705 min.y = MagickMin(min.y,p.y); \
1706 max.y = MagickMax(max.y,p.y); \
1711 case AffineDistortion:
1712 { double inverse[6];
1713 InvertAffineCoefficients(coeff, inverse);
1714 s.x = (double) image->page.x;
1715 s.y = (double) image->page.y;
1716 d.x = inverse[0]*s.x+inverse[1]*s.y+inverse[2];
1717 d.y = inverse[3]*s.x+inverse[4]*s.y+inverse[5];
1719 s.x = (double) image->page.x+image->columns;
1720 s.y = (double) image->page.y;
1721 d.x = inverse[0]*s.x+inverse[1]*s.y+inverse[2];
1722 d.y = inverse[3]*s.x+inverse[4]*s.y+inverse[5];
1724 s.x = (double) image->page.x;
1725 s.y = (double) image->page.y+image->rows;
1726 d.x = inverse[0]*s.x+inverse[1]*s.y+inverse[2];
1727 d.y = inverse[3]*s.x+inverse[4]*s.y+inverse[5];
1729 s.x = (double) image->page.x+image->columns;
1730 s.y = (double) image->page.y+image->rows;
1731 d.x = inverse[0]*s.x+inverse[1]*s.y+inverse[2];
1732 d.y = inverse[3]*s.x+inverse[4]*s.y+inverse[5];
1736 case PerspectiveDistortion:
1737 { double inverse[8], scale;
1738 InvertPerspectiveCoefficients(coeff, inverse);
1739 s.x = (double) image->page.x;
1740 s.y = (double) image->page.y;
1741 scale=inverse[6]*s.x+inverse[7]*s.y+1.0;
1742 scale=1.0/( (fabs(scale) <= MagickEpsilon) ? 1.0 : scale );
1743 d.x = scale*(inverse[0]*s.x+inverse[1]*s.y+inverse[2]);
1744 d.y = scale*(inverse[3]*s.x+inverse[4]*s.y+inverse[5]);
1746 s.x = (double) image->page.x+image->columns;
1747 s.y = (double) image->page.y;
1748 scale=inverse[6]*s.x+inverse[7]*s.y+1.0;
1749 scale=1.0/( (fabs(scale) <= MagickEpsilon) ? 1.0 : scale );
1750 d.x = scale*(inverse[0]*s.x+inverse[1]*s.y+inverse[2]);
1751 d.y = scale*(inverse[3]*s.x+inverse[4]*s.y+inverse[5]);
1753 s.x = (double) image->page.x;
1754 s.y = (double) image->page.y+image->rows;
1755 scale=inverse[6]*s.x+inverse[7]*s.y+1.0;
1756 scale=1.0/( (fabs(scale) <= MagickEpsilon) ? 1.0 : scale );
1757 d.x = scale*(inverse[0]*s.x+inverse[1]*s.y+inverse[2]);
1758 d.y = scale*(inverse[3]*s.x+inverse[4]*s.y+inverse[5]);
1760 s.x = (double) image->page.x+image->columns;
1761 s.y = (double) image->page.y+image->rows;
1762 scale=inverse[6]*s.x+inverse[7]*s.y+1.0;
1763 scale=1.0/( (fabs(scale) <= MagickEpsilon) ? 1.0 : scale );
1764 d.x = scale*(inverse[0]*s.x+inverse[1]*s.y+inverse[2]);
1765 d.y = scale*(inverse[3]*s.x+inverse[4]*s.y+inverse[5]);
1771 /* Forward Map Corners */
1772 a = coeff[0]-coeff[1]/2; ca = cos(a); sa = sin(a);
1776 d.x = (coeff[2]-coeff[3])*ca;
1777 d.y = (coeff[2]-coeff[3])*sa;
1779 a = coeff[0]+coeff[1]/2; ca = cos(a); sa = sin(a);
1783 d.x = (coeff[2]-coeff[3])*ca;
1784 d.y = (coeff[2]-coeff[3])*sa;
1786 /* Orthogonal points along top of arc */
1787 for( a=(double) (ceil((double) ((coeff[0]-coeff[1]/2.0)/MagickPI2))*MagickPI2);
1788 a<(coeff[0]+coeff[1]/2.0); a+=MagickPI2 ) {
1789 ca = cos(a); sa = sin(a);
1795 Convert the angle_to_width and radius_to_height
1796 to appropriate scaling factors, to allow faster processing
1797 in the mapping function.
1799 coeff[1] = (double) (Magick2PI*image->columns/coeff[1]);
1800 coeff[3] = (double)image->rows/coeff[3];
1803 case PolarDistortion:
1805 if (number_arguments < 2)
1806 coeff[2] = coeff[3] = 0.0;
1807 min.x = coeff[2]-coeff[0];
1808 max.x = coeff[2]+coeff[0];
1809 min.y = coeff[3]-coeff[0];
1810 max.y = coeff[3]+coeff[0];
1811 /* should be about 1.0 if Rmin = 0 */
1812 coeff[7]=(double) geometry.height/(coeff[0]-coeff[1]);
1815 case DePolarDistortion:
1817 /* direct calculation as it needs to tile correctly
1818 * for reversibility in a DePolar-Polar cycle */
1819 fix_bounds = MagickFalse;
1820 geometry.x = geometry.y = 0;
1821 geometry.height = (size_t) ceil(coeff[0]-coeff[1]);
1822 geometry.width = (size_t)
1823 ceil((coeff[0]-coeff[1])*(coeff[5]-coeff[4])*0.5);
1824 /* correct scaling factors relative to new size */
1825 coeff[6]=(coeff[5]-coeff[4])/geometry.width; /* changed width */
1826 coeff[7]=(coeff[0]-coeff[1])/geometry.height; /* should be about 1.0 */
1829 case Cylinder2PlaneDistortion:
1831 /* direct calculation so center of distortion is either a pixel
1832 * center, or pixel edge. This allows for reversibility of the
1834 geometry.x = geometry.y = 0;
1835 geometry.width = (size_t) ceil( 2.0*coeff[1]*tan(coeff[0]/2.0) );
1836 geometry.height = (size_t) ceil( 2.0*coeff[3]/cos(coeff[0]/2.0) );
1837 /* correct center of distortion relative to new size */
1838 coeff[4] = (double) geometry.width/2.0;
1839 coeff[5] = (double) geometry.height/2.0;
1840 fix_bounds = MagickFalse;
1843 case Plane2CylinderDistortion:
1845 /* direct calculation center is either pixel center, or pixel edge
1846 * so as to allow reversibility of the image distortion */
1847 geometry.x = geometry.y = 0;
1848 geometry.width = (size_t) ceil(coeff[0]*coeff[1]); /* FOV * radius */
1849 geometry.height = (size_t) (2*coeff[3]); /* input image height */
1850 /* correct center of distortion relative to new size */
1851 coeff[4] = (double) geometry.width/2.0;
1852 coeff[5] = (double) geometry.height/2.0;
1853 fix_bounds = MagickFalse;
1856 case ShepardsDistortion:
1857 case BilinearForwardDistortion:
1858 case BilinearReverseDistortion:
1860 case QuadrilateralDistortion:
1862 case PolynomialDistortion:
1863 case BarrelDistortion:
1864 case BarrelInverseDistortion:
1866 /* no calculated bestfit available for these distortions */
1867 bestfit = MagickFalse;
1868 fix_bounds = MagickFalse;
1872 /* Set the output image geometry to calculated 'bestfit'.
1873 Yes this tends to 'over do' the file image size, ON PURPOSE!
1874 Do not do this for DePolar which needs to be exact for virtual tiling.
1877 geometry.x = (ssize_t) floor(min.x-0.5);
1878 geometry.y = (ssize_t) floor(min.y-0.5);
1879 geometry.width=(size_t) ceil(max.x-geometry.x+0.5);
1880 geometry.height=(size_t) ceil(max.y-geometry.y+0.5);
1883 } /* end bestfit destination image calculations */
1885 /* The user provided a 'viewport' expert option which may
1886 overrides some parts of the current output image geometry.
1887 This also overrides its default 'bestfit' setting.
1889 { const char *artifact=GetImageArtifact(image,"distort:viewport");
1890 viewport_given = MagickFalse;
1891 if ( artifact != (const char *) NULL ) {
1892 (void) ParseAbsoluteGeometry(artifact,&geometry);
1893 viewport_given = MagickTrue;
1897 /* Verbose output */
1898 if ( GetImageArtifact(image,"verbose") != (const char *) NULL ) {
1901 char image_gen[MaxTextExtent];
1904 /* Set destination image size and virtual offset */
1905 if ( bestfit || viewport_given ) {
1906 (void) FormatLocaleString(image_gen, MaxTextExtent," -size %.20gx%.20g "
1907 "-page %+.20g%+.20g xc: +insert \\\n",(double) geometry.width,
1908 (double) geometry.height,(double) geometry.x,(double) geometry.y);
1909 lookup="v.p{ xx-v.page.x-.5, yy-v.page.y-.5 }";
1912 image_gen[0] = '\0'; /* no destination to generate */
1913 lookup = "p{ xx-page.x-.5, yy-page.y-.5 }"; /* simplify lookup */
1917 case AffineDistortion:
1921 inverse = (double *) AcquireQuantumMemory(6,sizeof(*inverse));
1922 if (inverse == (double *) NULL) {
1923 coeff = (double *) RelinquishMagickMemory(coeff);
1924 (void) ThrowMagickException(exception,GetMagickModule(),
1925 ResourceLimitError,"MemoryAllocationFailed",
1926 "%s", "DistortImages");
1927 return((Image *) NULL);
1929 InvertAffineCoefficients(coeff, inverse);
1930 CoefficientsToAffineArgs(inverse);
1931 (void) FormatLocaleFile(stderr, "Affine Projection:\n");
1932 (void) FormatLocaleFile(stderr, " -distort AffineProjection \\\n '");
1933 for (i=0; i < 5; i++)
1934 (void) FormatLocaleFile(stderr, "%lf,", inverse[i]);
1935 (void) FormatLocaleFile(stderr, "%lf'\n", inverse[5]);
1936 inverse = (double *) RelinquishMagickMemory(inverse);
1938 (void) FormatLocaleFile(stderr, "Affine Distort, FX Equivelent:\n");
1939 (void) FormatLocaleFile(stderr, "%s", image_gen);
1940 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x+0.5; jj=j+page.y+0.5;\n");
1941 (void) FormatLocaleFile(stderr, " xx=%+lf*ii %+lf*jj %+lf;\n",
1942 coeff[0], coeff[1], coeff[2]);
1943 (void) FormatLocaleFile(stderr, " yy=%+lf*ii %+lf*jj %+lf;\n",
1944 coeff[3], coeff[4], coeff[5]);
1945 (void) FormatLocaleFile(stderr, " %s' \\\n", lookup);
1950 case PerspectiveDistortion:
1954 inverse = (double *) AcquireQuantumMemory(8,sizeof(*inverse));
1955 if (inverse == (double *) NULL) {
1956 coeff = (double *) RelinquishMagickMemory(coeff);
1957 (void) ThrowMagickException(exception,GetMagickModule(),
1958 ResourceLimitError,"MemoryAllocationFailed",
1959 "%s", "DistortCoefficients");
1960 return((Image *) NULL);
1962 InvertPerspectiveCoefficients(coeff, inverse);
1963 (void) FormatLocaleFile(stderr, "Perspective Projection:\n");
1964 (void) FormatLocaleFile(stderr, " -distort PerspectiveProjection \\\n '");
1966 (void) FormatLocaleFile(stderr, "%lf, ", inverse[i]);
1967 (void) FormatLocaleFile(stderr, "\n ");
1969 (void) FormatLocaleFile(stderr, "%lf, ", inverse[i]);
1970 (void) FormatLocaleFile(stderr, "%lf'\n", inverse[7]);
1971 inverse = (double *) RelinquishMagickMemory(inverse);
1973 (void) FormatLocaleFile(stderr, "Perspective Distort, FX Equivelent:\n");
1974 (void) FormatLocaleFile(stderr, "%s", image_gen);
1975 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x+0.5; jj=j+page.y+0.5;\n");
1976 (void) FormatLocaleFile(stderr, " rr=%+lf*ii %+lf*jj + 1;\n",
1977 coeff[6], coeff[7]);
1978 (void) FormatLocaleFile(stderr, " xx=(%+lf*ii %+lf*jj %+lf)/rr;\n",
1979 coeff[0], coeff[1], coeff[2]);
1980 (void) FormatLocaleFile(stderr, " yy=(%+lf*ii %+lf*jj %+lf)/rr;\n",
1981 coeff[3], coeff[4], coeff[5]);
1982 (void) FormatLocaleFile(stderr, " rr%s0 ? %s : blue' \\\n",
1983 coeff[8] < 0 ? "<" : ">", lookup);
1987 case BilinearForwardDistortion:
1988 (void) FormatLocaleFile(stderr, "BilinearForward Mapping Equations:\n");
1989 (void) FormatLocaleFile(stderr, "%s", image_gen);
1990 (void) FormatLocaleFile(stderr, " i = %+lf*x %+lf*y %+lf*x*y %+lf;\n",
1991 coeff[0], coeff[1], coeff[2], coeff[3]);
1992 (void) FormatLocaleFile(stderr, " j = %+lf*x %+lf*y %+lf*x*y %+lf;\n",
1993 coeff[4], coeff[5], coeff[6], coeff[7]);
1996 (void) FormatLocaleFile(stderr, " c8 = %+lf c9 = 2*a = %+lf;\n",
1997 coeff[8], coeff[9]);
1999 (void) FormatLocaleFile(stderr, "BilinearForward Distort, FX Equivelent:\n");
2000 (void) FormatLocaleFile(stderr, "%s", image_gen);
2001 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x%+lf; jj=j+page.y%+lf;\n",
2002 0.5-coeff[3], 0.5-coeff[7]);
2003 (void) FormatLocaleFile(stderr, " bb=%lf*ii %+lf*jj %+lf;\n",
2004 coeff[6], -coeff[2], coeff[8]);
2005 /* Handle Special degenerate (non-quadratic) or trapezoidal case */
2006 if ( coeff[9] != 0 ) {
2007 (void) FormatLocaleFile(stderr, " rt=bb*bb %+lf*(%lf*ii%+lf*jj);\n",
2008 -2*coeff[9], coeff[4], -coeff[0]);
2009 (void) FormatLocaleFile(stderr, " yy=( -bb + sqrt(rt) ) / %lf;\n",
2012 (void) FormatLocaleFile(stderr, " yy=(%lf*ii%+lf*jj)/bb;\n",
2013 -coeff[4], coeff[0]);
2014 (void) FormatLocaleFile(stderr, " xx=(ii %+lf*yy)/(%lf %+lf*yy);\n",
2015 -coeff[1], coeff[0], coeff[2]);
2016 if ( coeff[9] != 0 )
2017 (void) FormatLocaleFile(stderr, " (rt < 0 ) ? red : %s'\n", lookup);
2019 (void) FormatLocaleFile(stderr, " %s' \\\n", lookup);
2022 case BilinearReverseDistortion:
2024 (void) FormatLocaleFile(stderr, "Polynomial Projection Distort:\n");
2025 (void) FormatLocaleFile(stderr, " -distort PolynomialProjection \\\n");
2026 (void) FormatLocaleFile(stderr, " '1.5, %lf, %lf, %lf, %lf,\n",
2027 coeff[3], coeff[0], coeff[1], coeff[2]);
2028 (void) FormatLocaleFile(stderr, " %lf, %lf, %lf, %lf'\n",
2029 coeff[7], coeff[4], coeff[5], coeff[6]);
2031 (void) FormatLocaleFile(stderr, "BilinearReverse Distort, FX Equivelent:\n");
2032 (void) FormatLocaleFile(stderr, "%s", image_gen);
2033 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x+0.5; jj=j+page.y+0.5;\n");
2034 (void) FormatLocaleFile(stderr, " xx=%+lf*ii %+lf*jj %+lf*ii*jj %+lf;\n",
2035 coeff[0], coeff[1], coeff[2], coeff[3]);
2036 (void) FormatLocaleFile(stderr, " yy=%+lf*ii %+lf*jj %+lf*ii*jj %+lf;\n",
2037 coeff[4], coeff[5], coeff[6], coeff[7]);
2038 (void) FormatLocaleFile(stderr, " %s' \\\n", lookup);
2041 case PolynomialDistortion:
2043 size_t nterms = (size_t) coeff[1];
2044 (void) FormatLocaleFile(stderr, "Polynomial (order %lg, terms %lu), FX Equivelent\n",
2045 coeff[0],(unsigned long) nterms);
2046 (void) FormatLocaleFile(stderr, "%s", image_gen);
2047 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x+0.5; jj=j+page.y+0.5;\n");
2048 (void) FormatLocaleFile(stderr, " xx =");
2049 for (i=0; i<(ssize_t) nterms; i++) {
2050 if ( i != 0 && i%4 == 0 ) (void) FormatLocaleFile(stderr, "\n ");
2051 (void) FormatLocaleFile(stderr, " %+lf%s", coeff[2+i],
2054 (void) FormatLocaleFile(stderr, ";\n yy =");
2055 for (i=0; i<(ssize_t) nterms; i++) {
2056 if ( i != 0 && i%4 == 0 ) (void) FormatLocaleFile(stderr, "\n ");
2057 (void) FormatLocaleFile(stderr, " %+lf%s", coeff[2+i+nterms],
2060 (void) FormatLocaleFile(stderr, ";\n %s' \\\n", lookup);
2065 (void) FormatLocaleFile(stderr, "Arc Distort, Internal Coefficients:\n");
2066 for ( i=0; i<5; i++ )
2067 (void) FormatLocaleFile(stderr, " c%.20g = %+lf\n", (double) i, coeff[i]);
2068 (void) FormatLocaleFile(stderr, "Arc Distort, FX Equivelent:\n");
2069 (void) FormatLocaleFile(stderr, "%s", image_gen);
2070 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x; jj=j+page.y;\n");
2071 (void) FormatLocaleFile(stderr, " xx=(atan2(jj,ii)%+lf)/(2*pi);\n",
2073 (void) FormatLocaleFile(stderr, " xx=xx-round(xx);\n");
2074 (void) FormatLocaleFile(stderr, " xx=xx*%lf %+lf;\n",
2075 coeff[1], coeff[4]);
2076 (void) FormatLocaleFile(stderr, " yy=(%lf - hypot(ii,jj)) * %lf;\n",
2077 coeff[2], coeff[3]);
2078 (void) FormatLocaleFile(stderr, " v.p{xx-.5,yy-.5}' \\\n");
2081 case PolarDistortion:
2083 (void) FormatLocaleFile(stderr, "Polar Distort, Internal Coefficents\n");
2084 for ( i=0; i<8; i++ )
2085 (void) FormatLocaleFile(stderr, " c%.20g = %+lf\n", (double) i, coeff[i]);
2086 (void) FormatLocaleFile(stderr, "Polar Distort, FX Equivelent:\n");
2087 (void) FormatLocaleFile(stderr, "%s", image_gen);
2088 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x%+lf; jj=j+page.y%+lf;\n",
2089 -coeff[2], -coeff[3]);
2090 (void) FormatLocaleFile(stderr, " xx=(atan2(ii,jj)%+lf)/(2*pi);\n",
2091 -(coeff[4]+coeff[5])/2 );
2092 (void) FormatLocaleFile(stderr, " xx=xx-round(xx);\n");
2093 (void) FormatLocaleFile(stderr, " xx=xx*2*pi*%lf + v.w/2;\n",
2095 (void) FormatLocaleFile(stderr, " yy=(hypot(ii,jj)%+lf)*%lf;\n",
2096 -coeff[1], coeff[7] );
2097 (void) FormatLocaleFile(stderr, " v.p{xx-.5,yy-.5}' \\\n");
2100 case DePolarDistortion:
2102 (void) FormatLocaleFile(stderr, "DePolar Distort, Internal Coefficents\n");
2103 for ( i=0; i<8; i++ )
2104 (void) FormatLocaleFile(stderr, " c%.20g = %+lf\n", (double) i, coeff[i]);
2105 (void) FormatLocaleFile(stderr, "DePolar Distort, FX Equivelent:\n");
2106 (void) FormatLocaleFile(stderr, "%s", image_gen);
2107 (void) FormatLocaleFile(stderr, " -fx 'aa=(i+.5)*%lf %+lf;\n", coeff[6], -coeff[4] );
2108 (void) FormatLocaleFile(stderr, " rr=(j+.5)*%lf %+lf;\n", coeff[7], +coeff[1] );
2109 (void) FormatLocaleFile(stderr, " xx=rr*sin(aa) %+lf;\n", coeff[2] );
2110 (void) FormatLocaleFile(stderr, " yy=rr*cos(aa) %+lf;\n", coeff[3] );
2111 (void) FormatLocaleFile(stderr, " v.p{xx-.5,yy-.5}' \\\n");
2114 case Cylinder2PlaneDistortion:
2116 (void) FormatLocaleFile(stderr, "Cylinder to Plane Distort, Internal Coefficents\n");
2117 (void) FormatLocaleFile(stderr, " cylinder_radius = %+lf\n", coeff[1]);
2118 (void) FormatLocaleFile(stderr, "Cylinder to Plane Distort, FX Equivelent:\n");
2119 (void) FormatLocaleFile(stderr, "%s", image_gen);
2120 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x%+lf+0.5; jj=j+page.y%+lf+0.5;\n",
2121 -coeff[4], -coeff[5]);
2122 (void) FormatLocaleFile(stderr, " aa=atan(ii/%+lf);\n", coeff[1] );
2123 (void) FormatLocaleFile(stderr, " xx=%lf*aa%+lf;\n",
2124 coeff[1], coeff[2] );
2125 (void) FormatLocaleFile(stderr, " yy=jj*cos(aa)%+lf;\n", coeff[3] );
2126 (void) FormatLocaleFile(stderr, " %s' \\\n", lookup);
2129 case Plane2CylinderDistortion:
2131 (void) FormatLocaleFile(stderr, "Plane to Cylinder Distort, Internal Coefficents\n");
2132 (void) FormatLocaleFile(stderr, " cylinder_radius = %+lf\n", coeff[1]);
2133 (void) FormatLocaleFile(stderr, "Plane to Cylinder Distort, FX Equivelent:\n");
2134 (void) FormatLocaleFile(stderr, "%s", image_gen);
2135 (void) FormatLocaleFile(stderr, " -fx 'ii=i+page.x%+lf+0.5; jj=j+page.y%+lf+0.5;\n",
2136 -coeff[4], -coeff[5]);
2137 (void) FormatLocaleFile(stderr, " ii=ii/%+lf;\n", coeff[1] );
2138 (void) FormatLocaleFile(stderr, " xx=%lf*tan(ii)%+lf;\n",
2139 coeff[1], coeff[2] );
2140 (void) FormatLocaleFile(stderr, " yy=jj/cos(ii)%+lf;\n",
2142 (void) FormatLocaleFile(stderr, " %s' \\\n", lookup);
2145 case BarrelDistortion:
2146 case BarrelInverseDistortion:
2148 /* NOTE: This does the barrel roll in pixel coords not image coords
2149 ** The internal distortion must do it in image coordinates,
2150 ** so that is what the center coeff (8,9) is given in.
2152 xc = ((double)image->columns-1.0)/2.0 + image->page.x;
2153 yc = ((double)image->rows-1.0)/2.0 + image->page.y;
2154 (void) FormatLocaleFile(stderr, "Barrel%s Distort, FX Equivelent:\n",
2155 method == BarrelDistortion ? "" : "Inv");
2156 (void) FormatLocaleFile(stderr, "%s", image_gen);
2157 if ( fabs(coeff[8]-xc-0.5) < 0.1 && fabs(coeff[9]-yc-0.5) < 0.1 )
2158 (void) FormatLocaleFile(stderr, " -fx 'xc=(w-1)/2; yc=(h-1)/2;\n");
2160 (void) FormatLocaleFile(stderr, " -fx 'xc=%lf; yc=%lf;\n",
2161 coeff[8]-0.5, coeff[9]-0.5);
2162 (void) FormatLocaleFile(stderr,
2163 " ii=i-xc; jj=j-yc; rr=hypot(ii,jj);\n");
2164 (void) FormatLocaleFile(stderr, " ii=ii%s(%lf*rr*rr*rr %+lf*rr*rr %+lf*rr %+lf);\n",
2165 method == BarrelDistortion ? "*" : "/",
2166 coeff[0],coeff[1],coeff[2],coeff[3]);
2167 (void) FormatLocaleFile(stderr, " jj=jj%s(%lf*rr*rr*rr %+lf*rr*rr %+lf*rr %+lf);\n",
2168 method == BarrelDistortion ? "*" : "/",
2169 coeff[4],coeff[5],coeff[6],coeff[7]);
2170 (void) FormatLocaleFile(stderr, " v.p{fx*ii+xc,fy*jj+yc}' \\\n");
2177 /* The user provided a 'scale' expert option will scale the
2178 output image size, by the factor given allowing for super-sampling
2179 of the distorted image space. Any scaling factors must naturally
2180 be halved as a result.
2182 { const char *artifact;
2183 artifact=GetImageArtifact(image,"distort:scale");
2184 output_scaling = 1.0;
2185 if (artifact != (const char *) NULL) {
2186 output_scaling = fabs(InterpretLocaleValue(artifact,(char **) NULL));
2187 geometry.width *= output_scaling;
2188 geometry.height *= output_scaling;
2189 geometry.x *= output_scaling;
2190 geometry.y *= output_scaling;
2191 if ( output_scaling < 0.1 ) {
2192 coeff = (double *) RelinquishMagickMemory(coeff);
2193 (void) ThrowMagickException(exception,GetMagickModule(),OptionError,
2194 "InvalidArgument","%s", "-set option:distort:scale" );
2195 return((Image *) NULL);
2197 output_scaling = 1/output_scaling;
2200 #define ScaleFilter(F,A,B,C,D) \
2201 ScaleResampleFilter( (F), \
2202 output_scaling*(A), output_scaling*(B), \
2203 output_scaling*(C), output_scaling*(D) )
2206 Initialize the distort image attributes.
2208 distort_image=CloneImage(image,geometry.width,geometry.height,MagickTrue,
2210 if (distort_image == (Image *) NULL)
2211 return((Image *) NULL);
2212 /* if image is ColorMapped - change it to DirectClass */
2213 if (SetImageStorageClass(distort_image,DirectClass) == MagickFalse)
2215 InheritException(exception,&distort_image->exception);
2216 distort_image=DestroyImage(distort_image);
2217 return((Image *) NULL);
2219 distort_image->page.x=geometry.x;
2220 distort_image->page.y=geometry.y;
2221 if (distort_image->background_color.alpha != OpaqueAlpha)
2222 distort_image->matte=MagickTrue;
2224 { /* ----- MAIN CODE -----
2225 Sample the source image to each pixel in the distort image.
2240 **restrict resample_filter;
2247 GetPixelInfo(distort_image,&zero);
2248 resample_filter=AcquireResampleFilterThreadSet(image,
2249 UndefinedVirtualPixelMethod,MagickFalse,exception);
2250 distort_view=AcquireCacheView(distort_image);
2251 #if defined(MAGICKCORE_OPENMP_SUPPORT)
2252 #pragma omp parallel for schedule(dynamic,4) shared(progress,status)
2254 for (j=0; j < (ssize_t) distort_image->rows; j++)
2257 id = GetOpenMPThreadId();
2260 validity; /* how mathematically valid is this the mapping */
2266 pixel, /* pixel color to assign to distorted image */
2267 invalid; /* the color to assign when distort result is invalid */
2271 s; /* transform destination image x,y to source image x,y */
2279 q=QueueCacheViewAuthenticPixels(distort_view,0,j,distort_image->columns,1,
2281 if (q == (const Quantum *) NULL)
2288 /* Define constant scaling vectors for Affine Distortions
2289 Other methods are either variable, or use interpolated lookup
2293 case AffineDistortion:
2294 ScaleFilter( resample_filter[id],
2296 coeff[3], coeff[4] );
2302 /* Initialize default pixel validity
2303 * negative: pixel is invalid output 'matte_color'
2304 * 0.0 to 1.0: antialiased, mix with resample output
2305 * 1.0 or greater: use resampled output.
2309 GetPixelInfo(distort_image,&invalid);
2310 SetPixelInfoPacket(distort_image,&distort_image->matte_color,&invalid);
2311 if (distort_image->colorspace == CMYKColorspace)
2312 ConvertRGBToCMYK(&invalid); /* what about other color spaces? */
2314 for (i=0; i < (ssize_t) distort_image->columns; i++)
2316 /* map pixel coordinate to distortion space coordinate */
2317 d.x = (double) (geometry.x+i+0.5)*output_scaling;
2318 d.y = (double) (geometry.y+j+0.5)*output_scaling;
2319 s = d; /* default is a no-op mapping */
2322 case AffineDistortion:
2324 s.x=coeff[0]*d.x+coeff[1]*d.y+coeff[2];
2325 s.y=coeff[3]*d.x+coeff[4]*d.y+coeff[5];
2326 /* Affine partial derivitives are constant -- set above */
2329 case PerspectiveDistortion:
2332 p,q,r,abs_r,abs_c6,abs_c7,scale;
2333 /* perspective is a ratio of affines */
2334 p=coeff[0]*d.x+coeff[1]*d.y+coeff[2];
2335 q=coeff[3]*d.x+coeff[4]*d.y+coeff[5];
2336 r=coeff[6]*d.x+coeff[7]*d.y+1.0;
2337 /* Pixel Validity -- is it a 'sky' or 'ground' pixel */
2338 validity = (r*coeff[8] < 0.0) ? 0.0 : 1.0;
2339 /* Determine horizon anti-alias blending */
2341 abs_c6 = fabs(coeff[6]);
2342 abs_c7 = fabs(coeff[7]);
2343 if ( abs_c6 > abs_c7 ) {
2344 if ( abs_r < abs_c6*output_scaling )
2345 validity = 0.5 - coeff[8]*r/(coeff[6]*output_scaling);
2347 else if ( abs_r < abs_c7*output_scaling )
2348 validity = 0.5 - coeff[8]*r/(coeff[7]*output_scaling);
2349 /* Perspective Sampling Point (if valid) */
2350 if ( validity > 0.0 ) {
2351 /* divide by r affine, for perspective scaling */
2355 /* Perspective Partial Derivatives or Scaling Vectors */
2357 ScaleFilter( resample_filter[id],
2358 (r*coeff[0] - p*coeff[6])*scale,
2359 (r*coeff[1] - p*coeff[7])*scale,
2360 (r*coeff[3] - q*coeff[6])*scale,
2361 (r*coeff[4] - q*coeff[7])*scale );
2365 case BilinearReverseDistortion:
2367 /* Reversed Mapped is just a simple polynomial */
2368 s.x=coeff[0]*d.x+coeff[1]*d.y+coeff[2]*d.x*d.y+coeff[3];
2369 s.y=coeff[4]*d.x+coeff[5]*d.y
2370 +coeff[6]*d.x*d.y+coeff[7];
2371 /* Bilinear partial derivitives of scaling vectors */
2372 ScaleFilter( resample_filter[id],
2373 coeff[0] + coeff[2]*d.y,
2374 coeff[1] + coeff[2]*d.x,
2375 coeff[4] + coeff[6]*d.y,
2376 coeff[5] + coeff[6]*d.x );
2379 case BilinearForwardDistortion:
2381 /* Forward mapped needs reversed polynomial equations
2382 * which unfortunatally requires a square root! */
2384 d.x -= coeff[3]; d.y -= coeff[7];
2385 b = coeff[6]*d.x - coeff[2]*d.y + coeff[8];
2386 c = coeff[4]*d.x - coeff[0]*d.y;
2389 /* Handle Special degenerate (non-quadratic) case
2390 * Currently without horizon anti-alising */
2391 if ( fabs(coeff[9]) < MagickEpsilon )
2394 c = b*b - 2*coeff[9]*c;
2398 s.y = ( -b + sqrt(c) )/coeff[9];
2400 if ( validity > 0.0 )
2401 s.x = ( d.x - coeff[1]*s.y) / ( coeff[0] + coeff[2]*s.y );
2403 /* NOTE: the sign of the square root should be -ve for parts
2404 where the source image becomes 'flipped' or 'mirrored'.
2405 FUTURE: Horizon handling
2406 FUTURE: Scaling factors or Deritives (how?)
2411 case BilinearDistortion:
2412 /* Bilinear mapping of any Quadrilateral to any Quadrilateral */
2413 /* UNDER DEVELOPMENT */
2416 case PolynomialDistortion:
2418 /* multi-ordered polynomial */
2423 nterms=(ssize_t)coeff[1];
2426 du,dv; /* the du,dv vectors from unit dx,dy -- derivatives */
2428 s.x=s.y=du.x=du.y=dv.x=dv.y=0.0;
2429 for(k=0; k < nterms; k++) {
2430 s.x += poly_basis_fn(k,d.x,d.y)*coeff[2+k];
2431 du.x += poly_basis_dx(k,d.x,d.y)*coeff[2+k];
2432 du.y += poly_basis_dy(k,d.x,d.y)*coeff[2+k];
2433 s.y += poly_basis_fn(k,d.x,d.y)*coeff[2+k+nterms];
2434 dv.x += poly_basis_dx(k,d.x,d.y)*coeff[2+k+nterms];
2435 dv.y += poly_basis_dy(k,d.x,d.y)*coeff[2+k+nterms];
2437 ScaleFilter( resample_filter[id], du.x,du.y,dv.x,dv.y );
2442 /* what is the angle and radius in the destination image */
2443 s.x = (double) ((atan2(d.y,d.x) - coeff[0])/Magick2PI);
2444 s.x -= MagickRound(s.x); /* angle */
2445 s.y = hypot(d.x,d.y); /* radius */
2447 /* Arc Distortion Partial Scaling Vectors
2448 Are derived by mapping the perpendicular unit vectors
2449 dR and dA*R*2PI rather than trying to map dx and dy
2450 The results is a very simple orthogonal aligned ellipse.
2452 if ( s.y > MagickEpsilon )
2453 ScaleFilter( resample_filter[id],
2454 (double) (coeff[1]/(Magick2PI*s.y)), 0, 0, coeff[3] );
2456 ScaleFilter( resample_filter[id],
2457 distort_image->columns*2, 0, 0, coeff[3] );
2459 /* now scale the angle and radius for source image lookup point */
2460 s.x = s.x*coeff[1] + coeff[4] + image->page.x +0.5;
2461 s.y = (coeff[2] - s.y) * coeff[3] + image->page.y;
2464 case PolarDistortion:
2465 { /* 2D Cartesain to Polar View */
2468 s.x = atan2(d.x,d.y) - (coeff[4]+coeff[5])/2;
2470 s.x -= MagickRound(s.x);
2471 s.x *= Magick2PI; /* angle - relative to centerline */
2472 s.y = hypot(d.x,d.y); /* radius */
2474 /* Polar Scaling vectors are based on mapping dR and dA vectors
2475 This results in very simple orthogonal scaling vectors
2477 if ( s.y > MagickEpsilon )
2478 ScaleFilter( resample_filter[id],
2479 (double) (coeff[6]/(Magick2PI*s.y)), 0, 0, coeff[7] );
2481 ScaleFilter( resample_filter[id],
2482 distort_image->columns*2, 0, 0, coeff[7] );
2484 /* now finish mapping radius/angle to source x,y coords */
2485 s.x = s.x*coeff[6] + (double)image->columns/2.0 + image->page.x;
2486 s.y = (s.y-coeff[1])*coeff[7] + image->page.y;
2489 case DePolarDistortion:
2490 { /* @D Polar to Carteasain */
2491 /* ignore all destination virtual offsets */
2492 d.x = ((double)i+0.5)*output_scaling*coeff[6]-coeff[4];
2493 d.y = ((double)j+0.5)*output_scaling*coeff[7]+coeff[1];
2494 s.x = d.y*sin(d.x) + coeff[2];
2495 s.y = d.y*cos(d.x) + coeff[3];
2496 /* derivatives are usless - better to use SuperSampling */
2499 case Cylinder2PlaneDistortion:
2500 { /* 3D Cylinder to Tangential Plane */
2502 /* relative to center of distortion */
2503 d.x -= coeff[4]; d.y -= coeff[5];
2504 d.x /= coeff[1]; /* x' = x/r */
2505 ax=atan(d.x); /* aa = atan(x/r) = u/r */
2506 cx=cos(ax); /* cx = cos(atan(x/r)) = 1/sqrt(x^2+u^2) */
2507 s.x = coeff[1]*ax; /* u = r*atan(x/r) */
2508 s.y = d.y*cx; /* v = y*cos(u/r) */
2509 /* derivatives... (see personnal notes) */
2510 ScaleFilter( resample_filter[id],
2511 1.0/(1.0+d.x*d.x), 0.0, -d.x*s.y*cx*cx/coeff[1], s.y/d.y );
2513 if ( i == 0 && j == 0 ) {
2514 fprintf(stderr, "x=%lf y=%lf u=%lf v=%lf\n", d.x*coeff[1], d.y, s.x, s.y);
2515 fprintf(stderr, "phi = %lf\n", (double)(ax * 180.0/MagickPI) );
2516 fprintf(stderr, "du/dx=%lf du/dx=%lf dv/dx=%lf dv/dy=%lf\n",
2517 1.0/(1.0+d.x*d.x), 0.0, -d.x*s.y*cx*cx/coeff[1], s.y/d.y );
2520 /* add center of distortion in source */
2521 s.x += coeff[2]; s.y += coeff[3];
2524 case Plane2CylinderDistortion:
2525 { /* 3D Cylinder to Tangential Plane */
2526 /* relative to center of distortion */
2527 d.x -= coeff[4]; d.y -= coeff[5];
2529 /* is pixel valid - horizon of a infinite Virtual-Pixel Plane
2530 * (see Anthony Thyssen's personal note) */
2531 validity = (double) (coeff[1]*MagickPI2 - fabs(d.x))/output_scaling + 0.5;
2533 if ( validity > 0.0 ) {
2535 d.x /= coeff[1]; /* x'= x/r */
2536 cx = 1/cos(d.x); /* cx = 1/cos(x/r) */
2537 tx = tan(d.x); /* tx = tan(x/r) */
2538 s.x = coeff[1]*tx; /* u = r * tan(x/r) */
2539 s.y = d.y*cx; /* v = y / cos(x/r) */
2540 /* derivatives... (see Anthony Thyssen's personal notes) */
2541 ScaleFilter( resample_filter[id],
2542 cx*cx, 0.0, s.y*cx/coeff[1], cx );
2544 /*if ( i == 0 && j == 0 )*/
2545 if ( d.x == 0.5 && d.y == 0.5 ) {
2546 fprintf(stderr, "x=%lf y=%lf u=%lf v=%lf\n", d.x*coeff[1], d.y, s.x, s.y);
2547 fprintf(stderr, "radius = %lf phi = %lf validity = %lf\n",
2548 coeff[1], (double)(d.x * 180.0/MagickPI), validity );
2549 fprintf(stderr, "du/dx=%lf du/dx=%lf dv/dx=%lf dv/dy=%lf\n",
2550 cx*cx, 0.0, s.y*cx/coeff[1], cx);
2554 /* add center of distortion in source */
2555 s.x += coeff[2]; s.y += coeff[3];
2558 case BarrelDistortion:
2559 case BarrelInverseDistortion:
2560 { /* Lens Barrel Distionion Correction */
2561 double r,fx,fy,gx,gy;
2562 /* Radial Polynomial Distortion (de-normalized) */
2565 r = sqrt(d.x*d.x+d.y*d.y);
2566 if ( r > MagickEpsilon ) {
2567 fx = ((coeff[0]*r + coeff[1])*r + coeff[2])*r + coeff[3];
2568 fy = ((coeff[4]*r + coeff[5])*r + coeff[6])*r + coeff[7];
2569 gx = ((3*coeff[0]*r + 2*coeff[1])*r + coeff[2])/r;
2570 gy = ((3*coeff[4]*r + 2*coeff[5])*r + coeff[6])/r;
2571 /* adjust functions and scaling for 'inverse' form */
2572 if ( method == BarrelInverseDistortion ) {
2573 fx = 1/fx; fy = 1/fy;
2574 gx *= -fx*fx; gy *= -fy*fy;
2576 /* Set the source pixel to lookup and EWA derivative vectors */
2577 s.x = d.x*fx + coeff[8];
2578 s.y = d.y*fy + coeff[9];
2579 ScaleFilter( resample_filter[id],
2580 gx*d.x*d.x + fx, gx*d.x*d.y,
2581 gy*d.x*d.y, gy*d.y*d.y + fy );
2584 /* Special handling to avoid divide by zero when r==0
2586 ** The source and destination pixels match in this case
2587 ** which was set at the top of the loop using s = d;
2588 ** otherwise... s.x=coeff[8]; s.y=coeff[9];
2590 if ( method == BarrelDistortion )
2591 ScaleFilter( resample_filter[id],
2592 coeff[3], 0, 0, coeff[7] );
2593 else /* method == BarrelInverseDistortion */
2594 /* FUTURE, trap for D==0 causing division by zero */
2595 ScaleFilter( resample_filter[id],
2596 1.0/coeff[3], 0, 0, 1.0/coeff[7] );
2600 case ShepardsDistortion:
2601 { /* Shepards Method, or Inverse Weighted Distance for
2602 displacement around the destination image control points
2603 The input arguments are the coefficents to the function.
2604 This is more of a 'displacement' function rather than an
2605 absolute distortion function.
2612 denominator = s.x = s.y = 0;
2613 for(i=0; i<number_arguments; i+=4) {
2615 ((double)d.x-arguments[i+2])*((double)d.x-arguments[i+2])
2616 + ((double)d.y-arguments[i+3])*((double)d.y-arguments[i+3]);
2622 s.x += (arguments[ i ]-arguments[i+2])*weight;
2623 s.y += (arguments[i+1]-arguments[i+3])*weight;
2624 denominator += weight;
2631 /* We can not determine derivatives using shepards method
2632 only color interpolatation, not area-resampling */
2636 break; /* use the default no-op given above */
2638 /* map virtual canvas location back to real image coordinate */
2639 if ( bestfit && method != ArcDistortion ) {
2640 s.x -= image->page.x;
2641 s.y -= image->page.y;
2646 if ( validity <= 0.0 ) {
2647 /* result of distortion is an invalid pixel - don't resample */
2648 SetPixelPixelInfo(distort_image,&invalid,q);
2651 /* resample the source image to find its correct color */
2652 (void) ResamplePixelColor(resample_filter[id],s.x,s.y,&pixel);
2653 /* if validity between 0.0 and 1.0 mix result with invalid pixel */
2654 if ( validity < 1.0 ) {
2655 /* Do a blend of sample color and invalid pixel */
2656 /* should this be a 'Blend', or an 'Over' compose */
2657 CompositePixelInfoBlend(&pixel,validity,&invalid,(1.0-validity),
2660 SetPixelPixelInfo(distort_image,&pixel,q);
2662 q+=GetPixelChannels(distort_image);
2664 sync=SyncCacheViewAuthenticPixels(distort_view,exception);
2665 if (sync == MagickFalse)
2667 if (image->progress_monitor != (MagickProgressMonitor) NULL)
2672 #if defined(MAGICKCORE_OPENMP_SUPPORT)
2673 #pragma omp critical (MagickCore_DistortImage)
2675 proceed=SetImageProgress(image,DistortImageTag,progress++,
2677 if (proceed == MagickFalse)
2681 distort_view=DestroyCacheView(distort_view);
2682 resample_filter=DestroyResampleFilterThreadSet(resample_filter);
2684 if (status == MagickFalse)
2685 distort_image=DestroyImage(distort_image);
2688 /* Arc does not return an offset unless 'bestfit' is in effect
2689 And the user has not provided an overriding 'viewport'.
2691 if ( method == ArcDistortion && !bestfit && !viewport_given ) {
2692 distort_image->page.x = 0;
2693 distort_image->page.y = 0;
2695 coeff = (double *) RelinquishMagickMemory(coeff);
2696 return(distort_image);
2700 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2704 % S p a r s e C o l o r I m a g e %
2708 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2710 % SparseColorImage(), given a set of coordinates, interpolates the colors
2711 % found at those coordinates, across the whole image, using various methods.
2713 % The format of the SparseColorImage() method is:
2715 % Image *SparseColorImage(const Image *image,const ChannelType channel,
2716 % const SparseColorMethod method,const size_t number_arguments,
2717 % const double *arguments,ExceptionInfo *exception)
2719 % A description of each parameter follows:
2721 % o image: the image to be filled in.
2723 % o channel: Specify which color values (in RGBKA sequence) are being set.
2724 % This also determines the number of color_values in above.
2726 % o method: the method to fill in the gradient between the control points.
2728 % The methods used for SparseColor() are often simular to methods
2729 % used for DistortImage(), and even share the same code for determination
2730 % of the function coefficents, though with more dimensions (or resulting
2733 % o number_arguments: the number of arguments given.
2735 % o arguments: array of floating point arguments for this method--
2736 % x,y,color_values-- with color_values given as normalized values.
2738 % o exception: return any errors or warnings in this structure
2741 MagickExport Image *SparseColorImage(const Image *image,
2742 const ChannelType channel,const SparseColorMethod method,
2743 const size_t number_arguments,const double *arguments,
2744 ExceptionInfo *exception)
2746 #define SparseColorTag "Distort/SparseColor"
2760 assert(image != (Image *) NULL);
2761 assert(image->signature == MagickSignature);
2762 if (image->debug != MagickFalse)
2763 (void) LogMagickEvent(TraceEvent,GetMagickModule(),"%s",image->filename);
2764 assert(exception != (ExceptionInfo *) NULL);
2765 assert(exception->signature == MagickSignature);
2767 /* Determine number of color values needed per control point */
2769 if ( channel & RedChannel ) number_colors++;
2770 if ( channel & GreenChannel ) number_colors++;
2771 if ( channel & BlueChannel ) number_colors++;
2772 if ( channel & AlphaChannel ) number_colors++;
2773 if ( channel & BlackChannel ) number_colors++;
2776 Convert input arguments into mapping coefficients, this this case
2777 we are mapping (distorting) colors, rather than coordinates.
2779 { DistortImageMethod
2782 distort_method=(DistortImageMethod) method;
2783 if ( distort_method >= SentinelDistortion )
2784 distort_method = ShepardsDistortion; /* Pretend to be Shepards */
2785 coeff = GenerateCoefficients(image, &distort_method, number_arguments,
2786 arguments, number_colors, exception);
2787 if ( coeff == (double *) NULL )
2788 return((Image *) NULL);
2790 Note some Distort Methods may fall back to other simpler methods,
2791 Currently the only fallback of concern is Bilinear to Affine
2792 (Barycentric), which is alaso sparse_colr method. This also ensures
2793 correct two and one color Barycentric handling.
2795 sparse_method = (SparseColorMethod) distort_method;
2796 if ( distort_method == ShepardsDistortion )
2797 sparse_method = method; /* return non-distiort methods to normal */
2800 /* Verbose output */
2801 if ( GetImageArtifact(image,"verbose") != (const char *) NULL ) {
2803 switch (sparse_method) {
2804 case BarycentricColorInterpolate:
2806 register ssize_t x=0;
2807 (void) FormatLocaleFile(stderr, "Barycentric Sparse Color:\n");
2808 if ( channel & RedChannel )
2809 (void) FormatLocaleFile(stderr, " -channel R -fx '%+lf*i %+lf*j %+lf' \\\n",
2810 coeff[x], coeff[x+1], coeff[x+2]),x+=3;
2811 if ( channel & GreenChannel )
2812 (void) FormatLocaleFile(stderr, " -channel G -fx '%+lf*i %+lf*j %+lf' \\\n",
2813 coeff[x], coeff[x+1], coeff[x+2]),x+=3;
2814 if ( channel & BlueChannel )
2815 (void) FormatLocaleFile(stderr, " -channel B -fx '%+lf*i %+lf*j %+lf' \\\n",
2816 coeff[x], coeff[x+1], coeff[x+2]),x+=3;
2817 if ( channel & BlackChannel )
2818 (void) FormatLocaleFile(stderr, " -channel K -fx '%+lf*i %+lf*j %+lf' \\\n",
2819 coeff[x], coeff[x+1], coeff[x+2]),x+=3;
2820 if ( channel & AlphaChannel )
2821 (void) FormatLocaleFile(stderr, " -channel A -fx '%+lf*i %+lf*j %+lf' \\\n",
2822 coeff[x], coeff[x+1], coeff[x+2]),x+=3;
2825 case BilinearColorInterpolate:
2827 register ssize_t x=0;
2828 (void) FormatLocaleFile(stderr, "Bilinear Sparse Color\n");
2829 if ( channel & RedChannel )
2830 (void) FormatLocaleFile(stderr, " -channel R -fx '%+lf*i %+lf*j %+lf*i*j %+lf;\n",
2831 coeff[ x ], coeff[x+1],
2832 coeff[x+2], coeff[x+3]),x+=4;
2833 if ( channel & GreenChannel )
2834 (void) FormatLocaleFile(stderr, " -channel G -fx '%+lf*i %+lf*j %+lf*i*j %+lf;\n",
2835 coeff[ x ], coeff[x+1],
2836 coeff[x+2], coeff[x+3]),x+=4;
2837 if ( channel & BlueChannel )
2838 (void) FormatLocaleFile(stderr, " -channel B -fx '%+lf*i %+lf*j %+lf*i*j %+lf;\n",
2839 coeff[ x ], coeff[x+1],
2840 coeff[x+2], coeff[x+3]),x+=4;
2841 if ( channel & BlackChannel )
2842 (void) FormatLocaleFile(stderr, " -channel K -fx '%+lf*i %+lf*j %+lf*i*j %+lf;\n",
2843 coeff[ x ], coeff[x+1],
2844 coeff[x+2], coeff[x+3]),x+=4;
2845 if ( channel & AlphaChannel )
2846 (void) FormatLocaleFile(stderr, " -channel A -fx '%+lf*i %+lf*j %+lf*i*j %+lf;\n",
2847 coeff[ x ], coeff[x+1],
2848 coeff[x+2], coeff[x+3]),x+=4;
2852 /* sparse color method is too complex for FX emulation */
2857 /* Generate new image for generated interpolated gradient.
2858 * ASIDE: Actually we could have just replaced the colors of the original
2859 * image, but IM Core policy, is if storage class could change then clone
2863 sparse_image=CloneImage(image,0,0,MagickTrue,exception);
2864 if (sparse_image == (Image *) NULL)
2865 return((Image *) NULL);
2866 if (SetImageStorageClass(sparse_image,DirectClass) == MagickFalse)
2867 { /* if image is ColorMapped - change it to DirectClass */
2868 InheritException(exception,&image->exception);
2869 sparse_image=DestroyImage(sparse_image);
2870 return((Image *) NULL);
2872 { /* ----- MAIN CODE ----- */
2887 sparse_view=AcquireCacheView(sparse_image);
2888 #if defined(MAGICKCORE_OPENMP_SUPPORT)
2889 #pragma omp parallel for schedule(dynamic,4) shared(progress,status)
2891 for (j=0; j < (ssize_t) sparse_image->rows; j++)
2897 pixel; /* pixel to assign to distorted image */
2905 q=GetCacheViewAuthenticPixels(sparse_view,0,j,sparse_image->columns,
2907 if (q == (const Quantum *) NULL)
2912 GetPixelInfo(sparse_image,&pixel);
2913 for (i=0; i < (ssize_t) image->columns; i++)
2915 SetPixelInfo(image,q,&pixel);
2916 switch (sparse_method)
2918 case BarycentricColorInterpolate:
2920 register ssize_t x=0;
2921 if ( channel & RedChannel )
2922 pixel.red = coeff[x]*i +coeff[x+1]*j
2924 if ( channel & GreenChannel )
2925 pixel.green = coeff[x]*i +coeff[x+1]*j
2927 if ( channel & BlueChannel )
2928 pixel.blue = coeff[x]*i +coeff[x+1]*j
2930 if ( channel & AlphaChannel )
2931 pixel.alpha = coeff[x]*i +coeff[x+1]*j
2933 if ( channel & BlackChannel )
2934 pixel.black = coeff[x]*i +coeff[x+1]*j
2938 case BilinearColorInterpolate:
2940 register ssize_t x=0;
2941 if ( channel & RedChannel )
2942 pixel.red = coeff[x]*i + coeff[x+1]*j +
2943 coeff[x+2]*i*j + coeff[x+3], x+=4;
2944 if ( channel & GreenChannel )
2945 pixel.green = coeff[x]*i + coeff[x+1]*j +
2946 coeff[x+2]*i*j + coeff[x+3], x+=4;
2947 if ( channel & BlueChannel )
2948 pixel.blue = coeff[x]*i + coeff[x+1]*j +
2949 coeff[x+2]*i*j + coeff[x+3], x+=4;
2950 if ( channel & AlphaChannel )
2951 pixel.alpha = coeff[x]*i + coeff[x+1]*j +
2952 coeff[x+2]*i*j + coeff[x+3], x+=4;
2953 if ( channel & BlackChannel )
2954 pixel.black = coeff[x]*i + coeff[x+1]*j +
2955 coeff[x+2]*i*j + coeff[x+3], x+=4;
2958 case InverseColorInterpolate:
2959 case ShepardsColorInterpolate:
2960 { /* Inverse (Squared) Distance weights average (IDW) */
2966 if ( channel & RedChannel ) pixel.red = 0.0;
2967 if ( channel & GreenChannel ) pixel.green = 0.0;
2968 if ( channel & BlueChannel ) pixel.blue = 0.0;
2969 if ( channel & BlackChannel ) pixel.black = 0.0;
2970 if ( channel & AlphaChannel ) pixel.alpha = 0.0;
2972 for(k=0; k<number_arguments; k+=2+number_colors) {
2973 register ssize_t x=(ssize_t) k+2;
2975 ((double)i-arguments[ k ])*((double)i-arguments[ k ])
2976 + ((double)j-arguments[k+1])*((double)j-arguments[k+1]);
2977 if ( method == InverseColorInterpolate )
2978 weight = sqrt(weight); /* inverse, not inverse squared */
2979 weight = ( weight < 1.0 ) ? 1.0 : 1.0/weight;
2980 if ( channel & RedChannel )
2981 pixel.red += arguments[x++]*weight;
2982 if ( channel & GreenChannel )
2983 pixel.green += arguments[x++]*weight;
2984 if ( channel & BlueChannel )
2985 pixel.blue += arguments[x++]*weight;
2986 if ( channel & AlphaChannel )
2987 pixel.alpha += arguments[x++]*weight;
2988 if ( channel & BlackChannel )
2989 pixel.black += arguments[x++]*weight;
2990 denominator += weight;
2992 if ( channel & RedChannel ) pixel.red /= denominator;
2993 if ( channel & GreenChannel ) pixel.green /= denominator;
2994 if ( channel & BlueChannel ) pixel.blue /= denominator;
2995 if ( channel & AlphaChannel ) pixel.alpha /= denominator;
2996 if ( channel & BlackChannel ) pixel.black /= denominator;
2999 case VoronoiColorInterpolate:
3001 { /* Just use the closest control point you can find! */
3005 minimum = MagickHuge;
3007 for(k=0; k<number_arguments; k+=2+number_colors) {
3009 ((double)i-arguments[ k ])*((double)i-arguments[ k ])
3010 + ((double)j-arguments[k+1])*((double)j-arguments[k+1]);
3011 if ( distance < minimum ) {
3012 register ssize_t x=(ssize_t) k+2;
3013 if ( channel & RedChannel ) pixel.red = arguments[x++];
3014 if ( channel & GreenChannel ) pixel.green = arguments[x++];
3015 if ( channel & BlueChannel ) pixel.blue = arguments[x++];
3016 if ( channel & AlphaChannel ) pixel.alpha = arguments[x++];
3017 if ( channel & BlackChannel ) pixel.black = arguments[x++];
3024 /* set the color directly back into the source image */
3025 if ( channel & RedChannel ) pixel.red *= QuantumRange;
3026 if ( channel & GreenChannel ) pixel.green *= QuantumRange;
3027 if ( channel & BlueChannel ) pixel.blue *= QuantumRange;
3028 if ( channel & BlackChannel ) pixel.black *= QuantumRange;
3029 if ( channel & AlphaChannel ) pixel.alpha *= QuantumRange;
3030 SetPixelPixelInfo(sparse_image,&pixel,q);
3031 q+=GetPixelChannels(sparse_image);
3033 sync=SyncCacheViewAuthenticPixels(sparse_view,exception);
3034 if (sync == MagickFalse)
3036 if (image->progress_monitor != (MagickProgressMonitor) NULL)
3041 #if defined(MAGICKCORE_OPENMP_SUPPORT)
3042 #pragma omp critical (MagickCore_SparseColorImage)
3044 proceed=SetImageProgress(image,SparseColorTag,progress++,image->rows);
3045 if (proceed == MagickFalse)
3049 sparse_view=DestroyCacheView(sparse_view);
3050 if (status == MagickFalse)
3051 sparse_image=DestroyImage(sparse_image);
3053 coeff = (double *) RelinquishMagickMemory(coeff);
3054 return(sparse_image);