| 1 |
– |
/* Copyright (c) 1994 Regents of the University of California */ |
| 2 |
– |
|
| 1 |
|
#ifndef lint |
| 2 |
< |
static char SCCSid[] = "$SunId$ LBL"; |
| 2 |
> |
static const char RCSid[] = "$Id$"; |
| 3 |
|
#endif |
| 6 |
– |
|
| 4 |
|
/* |
| 5 |
|
* Neural-Net quantization algorithm based on work of Anthony Dekker |
| 6 |
|
*/ |
| 48 |
|
|
| 49 |
|
#define setskip(sp,n) ((sp)[0]=(n)>>16,(sp)[1]=((n)>>8)&255,(sp)[2]=(n)&255) |
| 50 |
|
|
| 51 |
+ |
static cpyclrtab(); |
| 52 |
|
|
| 53 |
+ |
|
| 54 |
|
neu_init(npixels) /* initialize our sample array */ |
| 55 |
|
long npixels; |
| 56 |
|
{ |
| 131 |
|
cpyclrtab(); |
| 132 |
|
inxbuild(); |
| 133 |
|
/* we're done with our samples */ |
| 134 |
< |
free((char *)thesamples); |
| 134 |
> |
free((void *)thesamples); |
| 135 |
|
/* reset dithering function */ |
| 136 |
|
neu_dith_colrs((BYTE *)NULL, (COLR *)NULL, 0); |
| 137 |
|
/* return new color table size */ |
| 171 |
|
|
| 172 |
|
if (n != N) { /* get error propogation array */ |
| 173 |
|
if (N) { |
| 174 |
< |
free((char *)cerr); |
| 174 |
> |
free((void *)cerr); |
| 175 |
|
cerr = NULL; |
| 176 |
|
} |
| 177 |
|
if (n) |
| 207 |
|
} |
| 208 |
|
|
| 209 |
|
/* The following was adapted and modified from the original (GW) */ |
| 210 |
+ |
|
| 211 |
+ |
/* cheater definitions (GW) */ |
| 212 |
+ |
#define thepicture thesamples |
| 213 |
+ |
#define lengthcount (nsamples*3) |
| 214 |
+ |
#define samplefac 1 |
| 215 |
+ |
|
| 216 |
|
/*----------------------------------------------------------------------*/ |
| 217 |
|
/* */ |
| 218 |
|
/* NeuQuant */ |
| 219 |
|
/* -------- */ |
| 220 |
|
/* */ |
| 221 |
< |
/* Copyright: Anthony Dekker, June 1994 */ |
| 221 |
> |
/* Copyright: Anthony Dekker, November 1994 */ |
| 222 |
|
/* */ |
| 223 |
|
/* This program performs colour quantization of graphics images (SUN */ |
| 224 |
|
/* raster files). It uses a Kohonen Neural Network. It produces */ |
| 251 |
|
/* Email: [email protected] */ |
| 252 |
|
/*----------------------------------------------------------------------*/ |
| 253 |
|
|
| 254 |
< |
#define bool int |
| 255 |
< |
#define false 0 |
| 256 |
< |
#define true 1 |
| 254 |
> |
#define bool int |
| 255 |
> |
#define false 0 |
| 256 |
> |
#define true 1 |
| 257 |
|
|
| 258 |
< |
#define initrad 32 |
| 259 |
< |
#define radiusdec 30 |
| 260 |
< |
#define alphadec 30 |
| 258 |
> |
/* network defs */ |
| 259 |
> |
#define netsize clrtabsiz /* number of colours - can change this */ |
| 260 |
> |
#define maxnetpos (netsize-1) |
| 261 |
> |
#define netbiasshift 4 /* bias for colour values */ |
| 262 |
> |
#define ncycles 100 /* no. of learning cycles */ |
| 263 |
|
|
| 264 |
|
/* defs for freq and bias */ |
| 265 |
< |
#define gammashift 10 |
| 266 |
< |
#define betashift gammashift |
| 267 |
< |
#define intbiasshift 16 |
| 268 |
< |
#define intbias (1<<intbiasshift) |
| 269 |
< |
#define gamma (1<<gammashift) |
| 270 |
< |
#define beta (intbias>>betashift) |
| 265 |
> |
#define intbiasshift 16 /* bias for fractions */ |
| 266 |
> |
#define intbias (((int) 1)<<intbiasshift) |
| 267 |
> |
#define gammashift 10 /* gamma = 1024 */ |
| 268 |
> |
#define gamma (((int) 1)<<gammashift) |
| 269 |
> |
#define betashift 10 |
| 270 |
> |
#define beta (intbias>>betashift) /* beta = 1/1024 */ |
| 271 |
|
#define betagamma (intbias<<(gammashift-betashift)) |
| 265 |
– |
#define gammaphi (intbias<<(gammashift-8)) |
| 272 |
|
|
| 273 |
< |
/* defs for rad and alpha */ |
| 274 |
< |
#define maxrad (initrad+1) |
| 275 |
< |
#define radiusbiasshift 6 |
| 276 |
< |
#define radiusbias (1<<radiusbiasshift) |
| 277 |
< |
#define initradius ((int) (initrad*radiusbias)) |
| 278 |
< |
#define alphabiasshift 10 |
| 279 |
< |
#define initalpha (1<<alphabiasshift) |
| 273 |
> |
/* defs for decreasing radius factor */ |
| 274 |
> |
#define initrad (256>>3) /* for 256 cols, radius starts */ |
| 275 |
> |
#define radiusbiasshift 6 /* at 32.0 biased by 6 bits */ |
| 276 |
> |
#define radiusbias (((int) 1)<<radiusbiasshift) |
| 277 |
> |
#define initradius (initrad*radiusbias) /* and decreases by a */ |
| 278 |
> |
#define radiusdec 30 /* factor of 1/30 each cycle */ |
| 279 |
> |
|
| 280 |
> |
/* defs for decreasing alpha factor */ |
| 281 |
> |
#define alphabiasshift 10 /* alpha starts at 1.0 */ |
| 282 |
> |
#define initalpha (((int) 1)<<alphabiasshift) |
| 283 |
> |
int alphadec; /* biased by 10 bits */ |
| 284 |
> |
|
| 285 |
> |
/* radbias and alpharadbias used for radpower calculation */ |
| 286 |
|
#define radbiasshift 8 |
| 287 |
< |
#define radbias (1<<radbiasshift) |
| 287 |
> |
#define radbias (((int) 1)<<radbiasshift) |
| 288 |
|
#define alpharadbshift (alphabiasshift+radbiasshift) |
| 289 |
< |
#define alpharadbias (1<<alpharadbshift) |
| 289 |
> |
#define alpharadbias (((int) 1)<<alpharadbshift) |
| 290 |
|
|
| 291 |
< |
/* other defs */ |
| 292 |
< |
#define netbiasshift 4 |
| 293 |
< |
#define funnyshift (intbiasshift-netbiasshift) |
| 294 |
< |
#define maxnetval ((256<<netbiasshift)-1) |
| 295 |
< |
#define ncycles 100 |
| 296 |
< |
#define jump1 499 /* prime */ |
| 285 |
< |
#define jump2 491 /* prime */ |
| 286 |
< |
#define jump3 487 /* any pic whose size was divisible by all */ |
| 287 |
< |
#define jump4 503 /* four primes would be simply enormous */ |
| 291 |
> |
/* four primes near 500 - assume no image has a length so large */ |
| 292 |
> |
/* that it is divisible by all four primes */ |
| 293 |
> |
#define prime1 499 |
| 294 |
> |
#define prime2 491 |
| 295 |
> |
#define prime3 487 |
| 296 |
> |
#define prime4 503 |
| 297 |
|
|
| 289 |
– |
/* cheater definitions (GW) */ |
| 290 |
– |
#define thepicture thesamples |
| 291 |
– |
#define lengthcount (nsamples*3) |
| 292 |
– |
#define samplefac 1 |
| 293 |
– |
|
| 298 |
|
typedef int pixel[4]; /* BGRc */ |
| 299 |
+ |
pixel network[256]; |
| 300 |
|
|
| 301 |
< |
static pixel network[256]; |
| 301 |
> |
int netindex[256]; /* for network lookup - really 256 */ |
| 302 |
|
|
| 303 |
< |
static int netindex[256]; |
| 303 |
> |
int bias [256]; /* bias and freq arrays for learning */ |
| 304 |
> |
int freq [256]; |
| 305 |
> |
int radpower[initrad]; /* radpower for precomputation */ |
| 306 |
|
|
| 300 |
– |
static int bias [256]; |
| 301 |
– |
static int freq [256]; |
| 302 |
– |
static int radpower[256]; /* actually need only go up to maxrad */ |
| 307 |
|
|
| 308 |
< |
/* fixed space overhead 256*4+256+256+256+256 words = 256*8 = 8kB */ |
| 308 |
> |
/* initialise network in range (0,0,0) to (255,255,255) */ |
| 309 |
|
|
| 310 |
< |
|
| 307 |
< |
static |
| 308 |
< |
initnet() |
| 310 |
> |
initnet() |
| 311 |
|
{ |
| 312 |
|
register int i; |
| 313 |
|
register int *p; |
| 314 |
|
|
| 315 |
< |
for (i=0; i<clrtabsiz; i++) { |
| 315 |
> |
for (i=0; i<netsize; i++) { |
| 316 |
|
p = network[i]; |
| 317 |
< |
p[0] = |
| 318 |
< |
p[1] = |
| 317 |
< |
p[2] = (i<<8) / clrtabsiz; |
| 318 |
< |
freq[i] = intbias/clrtabsiz; /* 1/256 */ |
| 317 |
> |
p[0] = p[1] = p[2] = (i << (netbiasshift+8))/netsize; |
| 318 |
> |
freq[i] = intbias/netsize; /* 1/netsize */ |
| 319 |
|
bias[i] = 0; |
| 320 |
|
} |
| 321 |
|
} |
| 322 |
|
|
| 323 |
|
|
| 324 |
< |
static |
| 324 |
> |
/* do after unbias - insertion sort of network and build netindex[0..255] */ |
| 325 |
> |
|
| 326 |
|
inxbuild() |
| 327 |
|
{ |
| 328 |
|
register int i,j,smallpos,smallval; |
| 329 |
|
register int *p,*q; |
| 330 |
< |
int start,previous; |
| 330 |
> |
int previouscol,startpos; |
| 331 |
|
|
| 332 |
< |
previous = 0; |
| 333 |
< |
start = 0; |
| 334 |
< |
for (i=0; i<clrtabsiz; i++) { |
| 332 |
> |
previouscol = 0; |
| 333 |
> |
startpos = 0; |
| 334 |
> |
for (i=0; i<netsize; i++) { |
| 335 |
|
p = network[i]; |
| 336 |
|
smallpos = i; |
| 337 |
|
smallval = p[1]; /* index on g */ |
| 338 |
< |
/* find smallest in i+1..clrtabsiz-1 */ |
| 339 |
< |
for (j=i+1; j<clrtabsiz; j++) { |
| 338 |
> |
/* find smallest in i..netsize-1 */ |
| 339 |
> |
for (j=i+1; j<netsize; j++) { |
| 340 |
|
q = network[j]; |
| 341 |
|
if (q[1] < smallval) { /* index on g */ |
| 342 |
|
smallpos = j; |
| 344 |
|
} |
| 345 |
|
} |
| 346 |
|
q = network[smallpos]; |
| 347 |
+ |
/* swap p (i) and q (smallpos) entries */ |
| 348 |
|
if (i != smallpos) { |
| 349 |
|
j = q[0]; q[0] = p[0]; p[0] = j; |
| 350 |
|
j = q[1]; q[1] = p[1]; p[1] = j; |
| 352 |
|
j = q[3]; q[3] = p[3]; p[3] = j; |
| 353 |
|
} |
| 354 |
|
/* smallval entry is now in position i */ |
| 355 |
< |
if (smallval != previous) { |
| 356 |
< |
netindex[previous] = (start+i)>>1; |
| 357 |
< |
for (j=previous+1; j<smallval; j++) netindex[j] = i; |
| 358 |
< |
previous = smallval; |
| 359 |
< |
start = i; |
| 355 |
> |
if (smallval != previouscol) { |
| 356 |
> |
netindex[previouscol] = (startpos+i)>>1; |
| 357 |
> |
for (j=previouscol+1; j<smallval; j++) netindex[j] = i; |
| 358 |
> |
previouscol = smallval; |
| 359 |
> |
startpos = i; |
| 360 |
|
} |
| 361 |
|
} |
| 362 |
< |
netindex[previous] = (start+clrtabsiz-1)>>1; |
| 363 |
< |
for (j=previous+1; j<clrtabsiz; j++) netindex[j] = clrtabsiz-1; |
| 362 |
> |
netindex[previouscol] = (startpos+maxnetpos)>>1; |
| 363 |
> |
for (j=previouscol+1; j<256; j++) netindex[j] = maxnetpos; /* really 256 */ |
| 364 |
|
} |
| 365 |
|
|
| 366 |
|
|
| 367 |
< |
static int |
| 366 |
< |
inxsearch(b,g,r) /* accepts real BGR values after net is unbiased */ |
| 367 |
> |
int inxsearch(b,g,r) /* accepts real BGR values after net is unbiased */ |
| 368 |
|
register int b,g,r; |
| 369 |
|
{ |
| 370 |
< |
register int i,j,best,x,y,bestd; |
| 370 |
> |
register int i,j,dist,a,bestd; |
| 371 |
|
register int *p; |
| 372 |
+ |
int best; |
| 373 |
|
|
| 374 |
|
bestd = 1000; /* biggest possible dist is 256*3 */ |
| 375 |
|
best = -1; |
| 376 |
|
i = netindex[g]; /* index on g */ |
| 377 |
< |
j = i-1; |
| 377 |
> |
j = i-1; /* start at netindex[g] and work outwards */ |
| 378 |
|
|
| 379 |
< |
while ((i<clrtabsiz) || (j>=0)) { |
| 380 |
< |
if (i<clrtabsiz) { |
| 379 |
> |
while ((i<netsize) || (j>=0)) { |
| 380 |
> |
if (i<netsize) { |
| 381 |
|
p = network[i]; |
| 382 |
< |
x = p[1] - g; /* inx key */ |
| 383 |
< |
if (x >= bestd) i = clrtabsiz; /* stop iter */ |
| 382 |
> |
dist = p[1] - g; /* inx key */ |
| 383 |
> |
if (dist >= bestd) i = netsize; /* stop iter */ |
| 384 |
|
else { |
| 385 |
|
i++; |
| 386 |
< |
if (x<0) x = -x; |
| 387 |
< |
y = p[0] - b; |
| 388 |
< |
if (y<0) y = -y; |
| 389 |
< |
x += y; |
| 390 |
< |
if (x<bestd) { |
| 391 |
< |
y = p[2] - r; |
| 392 |
< |
if (y<0) y = -y; |
| 391 |
< |
x += y; /* x holds distance */ |
| 392 |
< |
if (x<bestd) {bestd=x; best=p[3];} |
| 386 |
> |
if (dist<0) dist = -dist; |
| 387 |
> |
a = p[0] - b; if (a<0) a = -a; |
| 388 |
> |
dist += a; |
| 389 |
> |
if (dist<bestd) { |
| 390 |
> |
a = p[2] - r; if (a<0) a = -a; |
| 391 |
> |
dist += a; |
| 392 |
> |
if (dist<bestd) {bestd=dist; best=p[3];} |
| 393 |
|
} |
| 394 |
|
} |
| 395 |
|
} |
| 396 |
|
if (j>=0) { |
| 397 |
|
p = network[j]; |
| 398 |
< |
x = g - p[1]; /* inx key - reverse dif */ |
| 399 |
< |
if (x >= bestd) j = -1; /* stop iter */ |
| 398 |
> |
dist = g - p[1]; /* inx key - reverse dif */ |
| 399 |
> |
if (dist >= bestd) j = -1; /* stop iter */ |
| 400 |
|
else { |
| 401 |
|
j--; |
| 402 |
< |
if (x<0) x = -x; |
| 403 |
< |
y = p[0] - b; |
| 404 |
< |
if (y<0) y = -y; |
| 405 |
< |
x += y; |
| 406 |
< |
if (x<bestd) { |
| 407 |
< |
y = p[2] - r; |
| 408 |
< |
if (y<0) y = -y; |
| 409 |
< |
x += y; /* x holds distance */ |
| 410 |
< |
if (x<bestd) {bestd=x; best=p[3];} |
| 402 |
> |
if (dist<0) dist = -dist; |
| 403 |
> |
a = p[0] - b; if (a<0) a = -a; |
| 404 |
> |
dist += a; |
| 405 |
> |
if (dist<bestd) { |
| 406 |
> |
a = p[2] - r; if (a<0) a = -a; |
| 407 |
> |
dist += a; |
| 408 |
> |
if (dist<bestd) {bestd=dist; best=p[3];} |
| 409 |
|
} |
| 410 |
|
} |
| 411 |
|
} |
| 414 |
|
} |
| 415 |
|
|
| 416 |
|
|
| 417 |
< |
static int |
| 418 |
< |
contest(b,g,r) /* accepts biased BGR values */ |
| 417 |
> |
/* finds closest neuron (min dist) and updates freq */ |
| 418 |
> |
/* finds best neuron (min dist-bias) and returns position */ |
| 419 |
> |
/* for frequently chosen neurons, freq[i] is high and bias[i] is negative */ |
| 420 |
> |
/* bias[i] = gamma*((1/netsize)-freq[i]) */ |
| 421 |
> |
|
| 422 |
> |
int contest(b,g,r) /* accepts biased BGR values */ |
| 423 |
|
register int b,g,r; |
| 424 |
|
{ |
| 425 |
< |
register int i,best,bestbias,x,y,bestd,bestbiasd; |
| 426 |
< |
register int *p,*q, *pp; |
| 425 |
> |
register int i,dist,a,biasdist,betafreq; |
| 426 |
> |
int bestpos,bestbiaspos,bestd,bestbiasd; |
| 427 |
> |
register int *p,*f, *n; |
| 428 |
|
|
| 429 |
< |
bestd = ~(1<<31); |
| 429 |
> |
bestd = ~(((int) 1)<<31); |
| 430 |
|
bestbiasd = bestd; |
| 431 |
< |
best = -1; |
| 432 |
< |
bestbias = best; |
| 433 |
< |
q = bias; |
| 434 |
< |
p = freq; |
| 435 |
< |
for (i=0; i<clrtabsiz; i++) { |
| 436 |
< |
pp = network[i]; |
| 437 |
< |
x = pp[0] - b; |
| 438 |
< |
if (x<0) x = -x; |
| 439 |
< |
y = pp[1] - g; |
| 440 |
< |
if (y<0) y = -y; |
| 441 |
< |
x += y; |
| 442 |
< |
y = pp[2] - r; |
| 443 |
< |
if (y<0) y = -y; |
| 444 |
< |
x += y; /* x holds distance */ |
| 445 |
< |
/* >> netbiasshift not needed if funnyshift used */ |
| 446 |
< |
if (x<bestd) {bestd=x; best=i;} |
| 447 |
< |
y = x - ((*q)>>funnyshift); /* y holds biasd */ |
| 448 |
< |
if (y<bestbiasd) {bestbiasd=y; bestbias=i;} |
| 446 |
< |
y = (*p >> betashift); /* y holds beta*freq */ |
| 447 |
< |
*p -= y; |
| 448 |
< |
*q += (y<<gammashift); |
| 449 |
< |
p++; |
| 450 |
< |
q++; |
| 431 |
> |
bestpos = -1; |
| 432 |
> |
bestbiaspos = bestpos; |
| 433 |
> |
p = bias; |
| 434 |
> |
f = freq; |
| 435 |
> |
|
| 436 |
> |
for (i=0; i<netsize; i++) { |
| 437 |
> |
n = network[i]; |
| 438 |
> |
dist = n[0] - b; if (dist<0) dist = -dist; |
| 439 |
> |
a = n[1] - g; if (a<0) a = -a; |
| 440 |
> |
dist += a; |
| 441 |
> |
a = n[2] - r; if (a<0) a = -a; |
| 442 |
> |
dist += a; |
| 443 |
> |
if (dist<bestd) {bestd=dist; bestpos=i;} |
| 444 |
> |
biasdist = dist - ((*p)>>(intbiasshift-netbiasshift)); |
| 445 |
> |
if (biasdist<bestbiasd) {bestbiasd=biasdist; bestbiaspos=i;} |
| 446 |
> |
betafreq = (*f >> betashift); |
| 447 |
> |
*f++ -= betafreq; |
| 448 |
> |
*p++ += (betafreq<<gammashift); |
| 449 |
|
} |
| 450 |
< |
freq[best] += beta; |
| 451 |
< |
bias[best] -= betagamma; |
| 452 |
< |
return(bestbias); |
| 450 |
> |
freq[bestpos] += beta; |
| 451 |
> |
bias[bestpos] -= betagamma; |
| 452 |
> |
return(bestbiaspos); |
| 453 |
|
} |
| 454 |
|
|
| 455 |
|
|
| 456 |
< |
static |
| 457 |
< |
alterneigh(rad,i,b,g,r) /* accepts biased BGR values */ |
| 456 |
> |
/* move neuron i towards (b,g,r) by factor alpha */ |
| 457 |
> |
|
| 458 |
> |
altersingle(alpha,i,b,g,r) /* accepts biased BGR values */ |
| 459 |
> |
register int alpha,i,b,g,r; |
| 460 |
> |
{ |
| 461 |
> |
register int *n; |
| 462 |
> |
|
| 463 |
> |
n = network[i]; /* alter hit neuron */ |
| 464 |
> |
*n -= (alpha*(*n - b)) / initalpha; |
| 465 |
> |
n++; |
| 466 |
> |
*n -= (alpha*(*n - g)) / initalpha; |
| 467 |
> |
n++; |
| 468 |
> |
*n -= (alpha*(*n - r)) / initalpha; |
| 469 |
> |
} |
| 470 |
> |
|
| 471 |
> |
|
| 472 |
> |
/* move neurons adjacent to i towards (b,g,r) by factor */ |
| 473 |
> |
/* alpha*(1-((i-j)^2/[r]^2)) precomputed as radpower[|i-j|]*/ |
| 474 |
> |
|
| 475 |
> |
alterneigh(rad,i,b,g,r) /* accents biased BGR values */ |
| 476 |
|
int rad,i; |
| 477 |
|
register int b,g,r; |
| 478 |
|
{ |
| 479 |
|
register int j,k,lo,hi,a; |
| 480 |
|
register int *p, *q; |
| 481 |
|
|
| 482 |
< |
lo = i-rad; |
| 483 |
< |
if (lo<-1) lo= -1; |
| 468 |
< |
hi = i+rad; |
| 469 |
< |
if (hi>clrtabsiz) hi=clrtabsiz; |
| 482 |
> |
lo = i-rad; if (lo<-1) lo= -1; |
| 483 |
> |
hi = i+rad; if (hi>netsize) hi=netsize; |
| 484 |
|
|
| 485 |
|
j = i+1; |
| 486 |
|
k = i-1; |
| 509 |
|
} |
| 510 |
|
|
| 511 |
|
|
| 498 |
– |
static |
| 499 |
– |
altersingle(alpha,j,b,g,r) /* accepts biased BGR values */ |
| 500 |
– |
register int alpha,j,b,g,r; |
| 501 |
– |
{ |
| 502 |
– |
register int *q; |
| 503 |
– |
|
| 504 |
– |
q = network[j]; /* alter hit neuron */ |
| 505 |
– |
*q -= (alpha*(*q - b)) / initalpha; |
| 506 |
– |
q++; |
| 507 |
– |
*q -= (alpha*(*q - g)) / initalpha; |
| 508 |
– |
q++; |
| 509 |
– |
*q -= (alpha*(*q - r)) / initalpha; |
| 510 |
– |
} |
| 511 |
– |
|
| 512 |
– |
|
| 513 |
– |
static |
| 512 |
|
learn() |
| 513 |
|
{ |
| 514 |
|
register int i,j,b,g,r; |
| 515 |
< |
int radius,rad,alpha,step,delta,upto; |
| 515 |
> |
int radius,rad,alpha,step,delta,samplepixels; |
| 516 |
|
register unsigned char *p; |
| 517 |
|
unsigned char *lim; |
| 518 |
|
|
| 519 |
< |
upto = lengthcount/(3*samplefac); |
| 522 |
< |
delta = upto/ncycles; |
| 523 |
< |
lim = thepicture + lengthcount; |
| 519 |
> |
alphadec = 30 + ((samplefac-1)/3); |
| 520 |
|
p = thepicture; |
| 521 |
+ |
lim = thepicture + lengthcount; |
| 522 |
+ |
samplepixels = lengthcount/(3*samplefac); |
| 523 |
+ |
delta = samplepixels/ncycles; |
| 524 |
|
alpha = initalpha; |
| 525 |
|
radius = initradius; |
| 526 |
+ |
|
| 527 |
|
rad = radius >> radiusbiasshift; |
| 528 |
|
if (rad <= 1) rad = 0; |
| 529 |
|
for (i=0; i<rad; i++) |
| 530 |
|
radpower[i] = alpha*(((rad*rad - i*i)*radbias)/(rad*rad)); |
| 531 |
< |
|
| 532 |
< |
if ((lengthcount%jump1) != 0) step = 3*jump1; |
| 531 |
> |
|
| 532 |
> |
if ((lengthcount%prime1) != 0) step = 3*prime1; |
| 533 |
|
else { |
| 534 |
< |
if ((lengthcount%jump2) !=0) step = 3*jump2; |
| 534 |
> |
if ((lengthcount%prime2) !=0) step = 3*prime2; |
| 535 |
|
else { |
| 536 |
< |
if ((lengthcount%jump3) !=0) step = 3*jump3; |
| 537 |
< |
else step = 3*jump4; |
| 536 |
> |
if ((lengthcount%prime3) !=0) step = 3*prime3; |
| 537 |
> |
else step = 3*prime4; |
| 538 |
|
} |
| 539 |
|
} |
| 540 |
+ |
|
| 541 |
|
i = 0; |
| 542 |
< |
while (i < upto) { |
| 542 |
> |
while (i < samplepixels) { |
| 543 |
|
b = p[0] << netbiasshift; |
| 544 |
|
g = p[1] << netbiasshift; |
| 545 |
|
r = p[2] << netbiasshift; |
| 546 |
|
j = contest(b,g,r); |
| 547 |
|
|
| 548 |
|
altersingle(alpha,j,b,g,r); |
| 549 |
< |
if (rad) alterneigh(rad,j,b,g,r); |
| 549 |
< |
/* alter neighbours */ |
| 549 |
> |
if (rad) alterneigh(rad,j,b,g,r); /* alter neighbours */ |
| 550 |
|
|
| 551 |
|
p += step; |
| 552 |
|
if (p >= lim) p -= lengthcount; |
| 563 |
|
} |
| 564 |
|
} |
| 565 |
|
|
| 566 |
< |
static |
| 566 |
> |
/* unbias network to give 0..255 entries */ |
| 567 |
> |
/* which can then be used for colour map */ |
| 568 |
> |
/* and record position i to prepare for sort */ |
| 569 |
> |
|
| 570 |
|
unbiasnet() |
| 571 |
|
{ |
| 572 |
|
int i,j; |
| 573 |
|
|
| 574 |
< |
for (i=0; i<clrtabsiz; i++) { |
| 574 |
> |
for (i=0; i<netsize; i++) { |
| 575 |
|
for (j=0; j<3; j++) |
| 576 |
|
network[i][j] >>= netbiasshift; |
| 577 |
|
network[i][3] = i; /* record colour no */ |
| 578 |
|
} |
| 579 |
|
} |
| 580 |
|
|
| 581 |
< |
/* Don't do this until the network has been unbiased */ |
| 581 |
> |
|
| 582 |
> |
/* Don't do this until the network has been unbiased (GW) */ |
| 583 |
|
|
| 584 |
|
static |
| 585 |
|
cpyclrtab() |
| 586 |
|
{ |
| 587 |
|
register int i,j,k; |
| 588 |
|
|
| 589 |
< |
for (j=0; j<clrtabsiz; j++) { |
| 589 |
> |
for (j=0; j<netsize; j++) { |
| 590 |
|
k = network[j][3]; |
| 591 |
|
for (i = 0; i < 3; i++) |
| 592 |
|
clrtab[k][i] = network[j][2-i]; |
