1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
|
# Copyright(c) 1986 Association of Universities for Research in Astronomy Inc.
include <imhdr.h>
include <error.h>
.help sigm2, sigm2_setup
.nf ___________________________________________________________________________
SIGM2 -- Get a line from a spatially scaled 2-dimensional image. This procedure
works like the regular IMIO get line procedure, but rescales the input
2-dimensional image in either or both axes upon input. If the magnification
ratio required is greater than 0 and less than 2 then linear interpolation is
used to resample the image. If the magnification ratio is greater than or
equal to 2 then the image is block averaged by the smallest factor which
reduces the magnification to the range 0-2 and then interpolated back up to
the desired size. In some cases this will smooth the data slightly, but the
operation is efficient and avoids aliasing effects.
si = sigm2_setup (im,pm, x1,x2,nx,xblk, y1,y2,ny,yblk, order)
sigm2_free (si)
ptr = sigm2[sr] (si, linenumber)
SIGM2_SETUP must be called to set up the transformations after mapping the
image and before performing any scaled i/o to the image. SIGM2_FREE must be
called when finished to return buffer space.
The SIGM routines are like SIGL routines except for the addition of
interpolation over bad pixels and order=-1 takes the maximum rather
than the average when doing block averaging or interpolation.
.endhelp ______________________________________________________________________
# Scaled image descriptor for 2-dim images
define SI_LEN 19
define SI_MAXDIM 2 # images of 2 dimensions supported
define SI_NBUFS 3 # nbuffers used by SIGL2
define SI_IM Memi[$1] # pointer to input image header
define SI_FP Memi[$1+1] # pointer to fixpix structure
define SI_GRID Memi[$1+2+$2-1] # pointer to array of X coords
define SI_NPIX Memi[$1+4+$2-1] # number of X coords
define SI_BAVG Memi[$1+6+$2-1] # X block averaging factor
define SI_INTERP Memi[$1+8+$2-1] # interpolate X axis
define SI_BUF Memi[$1+10+$2-1]# line buffers
define SI_BUFY Memi[$1+13+$2-1]# Y values of buffers
define SI_ORDER Memi[$1+15] # interpolator order
define SI_TYBUF Memi[$1+16] # buffer type
define SI_XOFF Memi[$1+17] # offset in input image to first X
define SI_INIT Memi[$1+18] # YES until first i/o is done
define OUTBUF SI_BUF($1,3)
define SI_TOL (1E-5) # close to a pixel
define INTVAL (abs ($1 - nint($1)) < SI_TOL)
define SWAPI {tempi=$2;$2=$1;$1=tempi}
define SWAPP {tempp=$2;$2=$1;$1=tempp}
define NOTSET (-9999)
# SIGM2_SETUP -- Set up the spatial transformation for SIGL2[SR]. Compute
# the block averaging factors (1 if no block averaging is required) and
# the sampling grid points, i.e., pixel coordinates of the output pixels in
# the input image.
pointer procedure sigm2_setup (im, pm, px1,px2,nx,xblk, py1,py2,ny,yblk, order)
pointer im # the input image
pointer pm # pixel mask
real px1, px2 # range in X to be sampled on an even grid
int nx # number of output pixels in X
int xblk # blocking factor in x
real py1, py2 # range in Y to be sampled on an even grid
int ny # number of output pixels in Y
int yblk # blocking factor in y
int order # interpolator order (0=replicate, 1=linear)
int npix, noldpix, nbavpix, i, j
int npts[SI_MAXDIM] # number of output points for axis
int blksize[SI_MAXDIM] # block averaging factor (npix per block)
real tau[SI_MAXDIM] # tau = p(i+1) - p(i) in fractional pixels
real p1[SI_MAXDIM] # starting pixel coords in each axis
real p2[SI_MAXDIM] # ending pixel coords in each axis
real scalar, start
pointer si, gp, xt_fpinit()
begin
iferr (call calloc (si, SI_LEN, TY_STRUCT))
call erract (EA_FATAL)
SI_IM(si) = im
SI_FP(si) = xt_fpinit (pm, 1, INDEFI)
SI_NPIX(si,1) = nx
SI_NPIX(si,2) = ny
SI_ORDER(si) = order
SI_INIT(si) = YES
p1[1] = px1 # X = index 1
p2[1] = px2
npts[1] = nx
blksize[1] = xblk
p1[2] = py1 # Y = index 2
p2[2] = py2
npts[2] = ny
blksize[2] = yblk
# Compute block averaging factors if not defined.
# If there is only one pixel then the block average is the average
# between the first and last point.
do i = 1, SI_MAXDIM {
if ((blksize[i] >= 1) && !IS_INDEFI (blksize[i])) {
if (npts[i] == 1)
tau[i] = 0.
else
tau[i] = (p2[i] - p1[i]) / (npts[i] - 1)
} else {
if (npts[i] == 1) {
tau[i] = 0.
blksize[i] = int (p2[i] - p1[i] + 1 + SI_TOL)
} else {
tau[i] = (p2[i] - p1[i]) / (npts[i] - 1)
if (tau[i] >= 2.0) {
# If nx or ny is not an integral multiple of the block
# averaging factor, noldpix is the next larger number
# which is an integral multiple. When the image is
# block averaged pixels will be replicated as necessary
# to fill the last block out to this size.
blksize[i] = int (tau[i] + SI_TOL)
npix = p2[i] - p1[i] + 1
noldpix = (npix+blksize[i]-1) / blksize[i] * blksize[i]
nbavpix = noldpix / blksize[i]
scalar = real (nbavpix - 1) / real (noldpix - 1)
p1[i] = (p1[i] - 1.0) * scalar + 1.0
p2[i] = (p2[i] - 1.0) * scalar + 1.0
tau[i] = (p2[i] - p1[i]) / (npts[i] - 1)
} else
blksize[i] = 1
}
}
}
SI_BAVG(si,1) = blksize[1]
SI_BAVG(si,2) = blksize[2]
# if (IS_INDEFI (xblk))
# xblk = blksize[1]
# if (IS_INDEFI (yblk))
# yblk = blksize[2]
# Allocate and initialize the grid arrays, specifying the X and Y
# coordinates of each pixel in the output image, in units of pixels
# in the input (possibly block averaged) image.
do i = 1, SI_MAXDIM {
# The X coordinate is special. We do not want to read entire
# input image lines if only a range of input X values are needed.
# Since the X grid vector passed to ALUI (the interpolator) must
# contain explicit offsets into the vector being interpolated,
# we must generate interpolator grid points starting near 1.0.
# The X origin, used to read the block averaged input line, is
# given by XOFF.
if (i == 1) {
SI_XOFF(si) = int (p1[i] + SI_TOL)
start = p1[1] - int (p1[i] + SI_TOL) + 1.0
} else
start = p1[i]
# Do the axes need to be interpolated?
if (INTVAL(start) && INTVAL(tau[i]))
SI_INTERP(si,i) = NO
else
SI_INTERP(si,i) = YES
# Allocate grid buffer and set the grid points.
iferr (call malloc (gp, npts[i], TY_REAL))
call erract (EA_FATAL)
SI_GRID(si,i) = gp
if (SI_ORDER(si) <= 0) {
do j = 0, npts[i]-1
Memr[gp+j] = int (start + (j * tau[i]) + 0.5 + SI_TOL)
} else {
do j = 0, npts[i]-1
Memr[gp+j] = start + (j * tau[i])
}
}
return (si)
end
# SIGM2_FREE -- Free storage associated with an image opened for scaled
# input. This does not close and unmap the image.
procedure sigm2_free (si)
pointer si
int i
begin
# Free fixpix structure.
call xt_fpfree (SI_FP(si))
# Free SIGM2 buffers.
do i = 1, SI_NBUFS
if (SI_BUF(si,i) != NULL)
call mfree (SI_BUF(si,i), SI_TYBUF(si))
# Free GRID buffers.
do i = 1, SI_MAXDIM
if (SI_GRID(si,i) != NULL)
call mfree (SI_GRID(si,i), TY_REAL)
call mfree (si, TY_STRUCT)
end
# SIGM2S -- Get a line of type short from a scaled image. Block averaging is
# done by a subprocedure; this procedure gets a line from a possibly block
# averaged image and if necessary interpolates it to the grid points of the
# output line.
pointer procedure sigm2s (si, lineno)
pointer si # pointer to SI descriptor
int lineno
pointer rawline, tempp, gp
int i, new_y[2], tempi, curbuf, altbuf
int nraw, npix, nblks_y, ybavg, x1, x2
real x, y, weight_1, weight_2
pointer si_blmavgs()
errchk si_blmavgs
begin
nraw = IM_LEN(SI_IM(si),1)
npix = SI_NPIX(si,1)
# Determine the range of X (in pixels on the block averaged input image)
# required for the interpolator.
gp = SI_GRID(si,1)
x1 = SI_XOFF(si)
x = Memr[gp+npix-1]
x2 = x1 + int(x)
if (INTVAL(x))
x2 = x2 - 1
x2 = max (x1 + 1, x2)
gp = SI_GRID(si,2)
y = Memr[gp+lineno-1]
# The following is an optimization provided for the case when it is
# not necessary to interpolate in either X or Y. Block averaging is
# permitted.
if (SI_INTERP(si,1) == NO && SI_INTERP(si,2) == NO)
return (si_blmavgs (SI_IM(si), SI_FP(si), x1, x2, int(y),
SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si)))
# If we are interpolating in Y two buffers are required, one for each
# of the two input image lines required to interpolate in Y. The lines
# stored in these buffers are interpolated in X to the output grid but
# not in Y. Both buffers are not required if we are not interpolating
# in Y, but we use them anyhow to simplify the code.
if (SI_INIT(si) == YES) {
do i = 1, 2 {
if (SI_BUF(si,i) != NULL)
call mfree (SI_BUF(si,i), SI_TYBUF(si))
call malloc (SI_BUF(si,i), npix, TY_SHORT)
SI_TYBUF(si) = TY_SHORT
SI_BUFY(si,i) = NOTSET
}
if (OUTBUF(si) != NULL)
call mfree (OUTBUF(si), SI_TYBUF(si))
call malloc (OUTBUF(si), npix, TY_SHORT)
SI_INIT(si) = NO
}
# If the Y value of the new line is not in range of the contents of the
# current line buffers, refill one or both buffers. To refill we must
# read a (possibly block averaged) input line and interpolate it onto
# the X grid. The X and Y values herein are in the coordinate system
# of the (possibly block averaged) input image.
new_y[1] = int(y)
new_y[2] = int(y) + 1
# Get the pair of lines whose integral Y values form an interval
# containing the fractional Y value of the output line. Sometimes the
# desired line will happen to be in the other buffer already, in which
# case we just have to swap buffers. Often the new line will be the
# current line, in which case nothing is done. This latter case occurs
# frequently when the magnification ratio is large.
curbuf = 1
altbuf = 2
do i = 1, 2 {
if (new_y[i] == SI_BUFY(si,i)) {
;
} else if (new_y[i] == SI_BUFY(si,altbuf)) {
SWAPP (SI_BUF(si,1), SI_BUF(si,2))
SWAPI (SI_BUFY(si,1), SI_BUFY(si,2))
} else {
# Get line and interpolate onto output grid. If interpolation
# is not required merely copy data out. This code is set up
# to always use two buffers; in effect, there is one buffer of
# look ahead, even when Y[i] is integral. This means that we
# will go out of bounds by one line at the top of the image.
# This is handled by copying the last line.
ybavg = SI_BAVG(si,2)
nblks_y = (IM_LEN (SI_IM(si), 2) + ybavg-1) / ybavg
if (new_y[i] <= nblks_y)
rawline = si_blmavgs (SI_IM(si), SI_FP(si), x1, x2,
new_y[i], SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si))
if (SI_INTERP(si,1) == NO) {
call amovs (Mems[rawline], Mems[SI_BUF(si,i)], npix)
} else if (SI_ORDER(si) == 0) {
call si_samples (Mems[rawline], Mems[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
} else if (SI_ORDER(si) == -1) {
call si_maxs (Mems[rawline], nraw,
Memr[SI_GRID(si,1)], Mems[SI_BUF(si,i)], npix)
} else {
call aluis (Mems[rawline], Mems[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
}
SI_BUFY(si,i) = new_y[i]
}
SWAPI (altbuf, curbuf)
}
# We now have two line buffers straddling the output Y value,
# interpolated to the X grid of the output line. To complete the
# bilinear interpolation operation we take a weighted sum of the two
# lines. If the range from SI_BUFY(si,1) to SI_BUFY(si,2) is repeatedly
# interpolated in Y no additional i/o occurs and the linear
# interpolation operation (ALUI) does not have to be repeated (only the
# weighted sum is required). If the distance of Y from one of the
# buffers is zero then we do not even have to take a weighted sum.
# This is not unusual because we may be called with a magnification
# of 1.0 in Y.
weight_1 = 1.0 - (y - SI_BUFY(si,1))
weight_2 = 1.0 - weight_1
if (weight_1 < SI_TOL)
return (SI_BUF(si,2))
else if (weight_2 < SI_TOL || SI_ORDER(si) == 0)
return (SI_BUF(si,1))
else if (SI_ORDER(si) == -1) {
call amaxs (Mems[SI_BUF(si,1)], Mems[SI_BUF(si,2)],
Mems[OUTBUF(si)], npix)
return (OUTBUF(si))
} else {
call awsus (Mems[SI_BUF(si,1)], Mems[SI_BUF(si,2)],
Mems[OUTBUF(si)], npix, weight_1, weight_2)
return (OUTBUF(si))
}
end
# SI_BLMAVGS -- Get a line from a block averaged image of type short.
# For example, block averaging by a factor of 2 means that pixels 1 and 2
# are averaged to produce the first output pixel, 3 and 4 are averaged to
# produce the second output pixel, and so on. If the length of an axis
# is not an integral multiple of the block size then the last pixel in the
# last block will be replicated to fill out the block; the average is still
# defined even if a block is not full.
pointer procedure si_blmavgs (im, fp, x1, x2, y, xbavg, ybavg, order)
pointer im # input image
pointer fp # fixpix structure
int x1, x2 # range of x blocks to be read
int y # y block to be read
int xbavg, ybavg # X and Y block averaging factors
int order # averaging option
real sum
short blkmax
pointer sp, a, b
int nblks_x, nblks_y, ncols, nlines, xoff, blk1, blk2, i, j, k
int first_line, nlines_in_sum, npix, nfull_blks, count
pointer xt_fps()
errchk xt_fps
begin
call smark (sp)
ncols = IM_LEN(im,1)
nlines = IM_LEN(im,2)
xoff = (x1 - 1) * xbavg + 1
npix = min (ncols, xoff + (x2 - x1 + 1) * xbavg - 1) - xoff + 1
if ((xbavg < 1) || (ybavg < 1))
call error (1, "si_blmavg: illegal block size")
else if (x1 < 1 || x2 > ncols)
call error (2, "si_blmavg: column index out of bounds")
else if ((xbavg == 1) && (ybavg == 1))
return (xt_fps (fp, im, y, NULL) + xoff - 1)
nblks_x = (npix + xbavg-1) / xbavg
nblks_y = (nlines + ybavg-1) / ybavg
if (y < 1 || y > nblks_y)
call error (2, "si_blmavg: block number out of range")
if (ybavg > 1) {
call salloc (b, nblks_x, TY_LONG)
call aclrl (Meml[b], nblks_x)
nlines_in_sum = 0
}
# Read and accumulate all input lines in the block.
first_line = (y - 1) * ybavg + 1
do i = first_line, min (nlines, first_line + ybavg - 1) {
# Get line from input image.
a = xt_fps (fp, im, i, NULL) + xoff - 1
# Block average line in X.
if (xbavg > 1) {
# First block average only the full blocks.
nfull_blks = npix / xbavg
if (order == -1) {
blk1 = a
do j = 1, nfull_blks {
blk2 = blk1 + xbavg
blkmax = Mems[blk1]
do k = blk1+1, blk2-1
blkmax = max (blkmax, Mems[k])
Mems[a+j-1] = blkmax
blk1 = blk2
}
} else
call abavs (Mems[a], Mems[a], nfull_blks, xbavg)
# Now average the final partial block, if any.
if (nfull_blks < nblks_x) {
if (order == -1) {
blkmax = Mems[blk1]
do k = blk1+1, a+npix-1
blkmax = max (blkmax, Mems[k])
Mems[a+j-1] = blkmax
} else {
sum = 0.0
count = 0
do j = nfull_blks * xbavg + 1, npix {
sum = sum + Mems[a+j-1]
count = count + 1
}
Mems[a+nblks_x-1] = sum / count
}
}
}
# Add line into block sum. Keep track of number of lines in sum
# so that we can compute block average later.
if (ybavg > 1) {
if (order == -1) {
do j = 0, nblks_x-1
Meml[b+j] = max (Meml[b+j], long (Mems[a+j]))
} else {
do j = 0, nblks_x-1
Meml[b+j] = Meml[b+j] + Mems[a+j]
nlines_in_sum = nlines_in_sum + 1
}
}
}
# Compute the block average in Y from the sum of all lines block
# averaged in X. Overwrite buffer A, the buffer returned by IMIO.
# This is kosher because the block averaged line is never longer
# than an input line.
if (ybavg > 1) {
if (order == -1) {
do i = 0, nblks_x-1
Mems[a+i] = Meml[b+i]
} else {
do i = 0, nblks_x-1
Mems[a+i] = Meml[b+i] / real(nlines_in_sum)
}
}
call sfree (sp)
return (a)
end
# SI_MAXS -- Resample a line via maximum value.
procedure si_maxs (a, na, x, b, nb)
short a[na] # input array
int na # input size
real x[nb] # sample grid
short b[nb] # output arrays
int nb # output size
int i
begin
do i = 1, nb
b[i] = max (a[int(x[i])], a[min(na,int(x[i]+1))])
end
# SIGM2I -- Get a line of type short from a scaled image. Block averaging is
# done by a subprocedure; this procedure gets a line from a possibly block
# averaged image and if necessary interpolates it to the grid points of the
# output line.
pointer procedure sigm2i (si, lineno)
pointer si # pointer to SI descriptor
int lineno
pointer rawline, tempp, gp
int i, new_y[2], tempi, curbuf, altbuf
int nraw, npix, nblks_y, ybavg, x1, x2
real x, y, weight_1, weight_2
pointer si_blmavgi()
errchk si_blmavgi
begin
nraw = IM_LEN(SI_IM(si),1)
npix = SI_NPIX(si,1)
# Determine the range of X (in pixels on the block averaged input image)
# required for the interpolator.
gp = SI_GRID(si,1)
x1 = SI_XOFF(si)
x = Memr[gp+npix-1]
x2 = x1 + int(x)
if (INTVAL(x))
x2 = x2 - 1
x2 = max (x1 + 1, x2)
gp = SI_GRID(si,2)
y = Memr[gp+lineno-1]
# The following is an optimization provided for the case when it is
# not necessary to interpolate in either X or Y. Block averaging is
# permitted.
if (SI_INTERP(si,1) == NO && SI_INTERP(si,2) == NO)
return (si_blmavgi (SI_IM(si), SI_FP(si), x1, x2, int(y),
SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si)))
# If we are interpolating in Y two buffers are required, one for each
# of the two input image lines required to interpolate in Y. The lines
# stored in these buffers are interpolated in X to the output grid but
# not in Y. Both buffers are not required if we are not interpolating
# in Y, but we use them anyhow to simplify the code.
if (SI_INIT(si) == YES) {
do i = 1, 2 {
if (SI_BUF(si,i) != NULL)
call mfree (SI_BUF(si,i), SI_TYBUF(si))
call malloc (SI_BUF(si,i), npix, TY_INT)
SI_TYBUF(si) = TY_INT
SI_BUFY(si,i) = NOTSET
}
if (OUTBUF(si) != NULL)
call mfree (OUTBUF(si), SI_TYBUF(si))
call malloc (OUTBUF(si), npix, TY_INT)
SI_INIT(si) = NO
}
# If the Y value of the new line is not in range of the contents of the
# current line buffers, refill one or both buffers. To refill we must
# read a (possibly block averaged) input line and interpolate it onto
# the X grid. The X and Y values herein are in the coordinate system
# of the (possibly block averaged) input image.
new_y[1] = int(y)
new_y[2] = int(y) + 1
# Get the pair of lines whose integral Y values form an interval
# containing the fractional Y value of the output line. Sometimes the
# desired line will happen to be in the other buffer already, in which
# case we just have to swap buffers. Often the new line will be the
# current line, in which case nothing is done. This latter case occurs
# frequently when the magnification ratio is large.
curbuf = 1
altbuf = 2
do i = 1, 2 {
if (new_y[i] == SI_BUFY(si,i)) {
;
} else if (new_y[i] == SI_BUFY(si,altbuf)) {
SWAPP (SI_BUF(si,1), SI_BUF(si,2))
SWAPI (SI_BUFY(si,1), SI_BUFY(si,2))
} else {
# Get line and interpolate onto output grid. If interpolation
# is not required merely copy data out. This code is set up
# to always use two buffers; in effect, there is one buffer of
# look ahead, even when Y[i] is integral. This means that we
# will go out of bounds by one line at the top of the image.
# This is handled by copying the last line.
ybavg = SI_BAVG(si,2)
nblks_y = (IM_LEN (SI_IM(si), 2) + ybavg-1) / ybavg
if (new_y[i] <= nblks_y)
rawline = si_blmavgi (SI_IM(si), SI_FP(si), x1, x2,
new_y[i], SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si))
if (SI_INTERP(si,1) == NO) {
call amovi (Memi[rawline], Memi[SI_BUF(si,i)], npix)
} else if (SI_ORDER(si) == 0) {
call si_samplei (Memi[rawline], Memi[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
} else if (SI_ORDER(si) == -1) {
call si_maxi (Memi[rawline], nraw,
Memr[SI_GRID(si,1)], Memi[SI_BUF(si,i)], npix)
} else {
call aluii (Memi[rawline], Memi[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
}
SI_BUFY(si,i) = new_y[i]
}
SWAPI (altbuf, curbuf)
}
# We now have two line buffers straddling the output Y value,
# interpolated to the X grid of the output line. To complete the
# bilinear interpolation operation we take a weighted sum of the two
# lines. If the range from SI_BUFY(si,1) to SI_BUFY(si,2) is repeatedly
# interpolated in Y no additional i/o occurs and the linear
# interpolation operation (ALUI) does not have to be repeated (only the
# weighted sum is required). If the distance of Y from one of the
# buffers is zero then we do not even have to take a weighted sum.
# This is not unusual because we may be called with a magnification
# of 1.0 in Y.
weight_1 = 1.0 - (y - SI_BUFY(si,1))
weight_2 = 1.0 - weight_1
if (weight_1 < SI_TOL)
return (SI_BUF(si,2))
else if (weight_2 < SI_TOL || SI_ORDER(si) == 0)
return (SI_BUF(si,1))
else if (SI_ORDER(si) == -1) {
call amaxi (Memi[SI_BUF(si,1)], Memi[SI_BUF(si,2)],
Memi[OUTBUF(si)], npix)
return (OUTBUF(si))
} else {
call awsui (Memi[SI_BUF(si,1)], Memi[SI_BUF(si,2)],
Memi[OUTBUF(si)], npix, weight_1, weight_2)
return (OUTBUF(si))
}
end
# SI_BLMAVGI -- Get a line from a block averaged image of type integer.
# For example, block averaging by a factor of 2 means that pixels 1 and 2
# are averaged to produce the first output pixel, 3 and 4 are averaged to
# produce the second output pixel, and so on. If the length of an axis
# is not an integral multiple of the block size then the last pixel in the
# last block will be replicated to fill out the block; the average is still
# defined even if a block is not full.
pointer procedure si_blmavgi (im, fp, x1, x2, y, xbavg, ybavg, order)
pointer im # input image
pointer fp # fixpix structure
int x1, x2 # range of x blocks to be read
int y # y block to be read
int xbavg, ybavg # X and Y block averaging factors
int order # averaging option
real sum
int blkmax
pointer sp, a, b
int nblks_x, nblks_y, ncols, nlines, xoff, blk1, blk2, i, j, k
int first_line, nlines_in_sum, npix, nfull_blks, count
pointer xt_fpi()
errchk xt_fpi
begin
call smark (sp)
ncols = IM_LEN(im,1)
nlines = IM_LEN(im,2)
xoff = (x1 - 1) * xbavg + 1
npix = min (ncols, xoff + (x2 - x1 + 1) * xbavg - 1) - xoff + 1
if ((xbavg < 1) || (ybavg < 1))
call error (1, "si_blmavg: illegal block size")
else if (x1 < 1 || x2 > ncols)
call error (2, "si_blmavg: column index out of bounds")
else if ((xbavg == 1) && (ybavg == 1))
return (xt_fpi (fp, im, y, NULL) + xoff - 1)
nblks_x = (npix + xbavg-1) / xbavg
nblks_y = (nlines + ybavg-1) / ybavg
if (y < 1 || y > nblks_y)
call error (2, "si_blmavg: block number out of range")
if (ybavg > 1) {
call salloc (b, nblks_x, TY_LONG)
call aclrl (Meml[b], nblks_x)
nlines_in_sum = 0
}
# Read and accumulate all input lines in the block.
first_line = (y - 1) * ybavg + 1
do i = first_line, min (nlines, first_line + ybavg - 1) {
# Get line from input image.
a = xt_fpi (fp, im, i, NULL) + xoff - 1
# Block average line in X.
if (xbavg > 1) {
# First block average only the full blocks.
nfull_blks = npix / xbavg
if (order == -1) {
blk1 = a
do j = 1, nfull_blks {
blk2 = blk1 + xbavg
blkmax = Memi[blk1]
do k = blk1+1, blk2-1
blkmax = max (blkmax, Memi[k])
Memi[a+j-1] = blkmax
blk1 = blk2
}
} else
call abavi (Memi[a], Memi[a], nfull_blks, xbavg)
# Now average the final partial block, if any.
if (nfull_blks < nblks_x) {
if (order == -1) {
blkmax = Memi[blk1]
do k = blk1+1, a+npix-1
blkmax = max (blkmax, Memi[k])
Memi[a+j-1] = blkmax
} else {
sum = 0.0
count = 0
do j = nfull_blks * xbavg + 1, npix {
sum = sum + Memi[a+j-1]
count = count + 1
}
Memi[a+nblks_x-1] = sum / count
}
}
}
# Add line into block sum. Keep track of number of lines in sum
# so that we can compute block average later.
if (ybavg > 1) {
if (order == -1) {
do j = 0, nblks_x-1
Meml[b+j] = max (Meml[b+j], long (Memi[a+j]))
} else {
do j = 0, nblks_x-1
Meml[b+j] = Meml[b+j] + Memi[a+j]
nlines_in_sum = nlines_in_sum + 1
}
}
}
# Compute the block average in Y from the sum of all lines block
# averaged in X. Overwrite buffer A, the buffer returned by IMIO.
# This is kosher because the block averaged line is never longer
# than an input line.
if (ybavg > 1) {
if (order == -1) {
do i = 0, nblks_x-1
Memi[a+i] = Meml[b+i]
} else {
do i = 0, nblks_x-1
Memi[a+i] = Meml[b+i] / real(nlines_in_sum)
}
}
call sfree (sp)
return (a)
end
# SI_MAXI -- Resample a line via maximum value.
procedure si_maxi (a, na, x, b, nb)
int a[na] # input array
int na # input size
real x[nb] # sample grid
int b[nb] # output arrays
int nb # output size
int i
begin
do i = 1, nb
b[i] = max (a[int(x[i])], a[min(na,int(x[i]+1))])
end
# SIGM2R -- Get a line of type real from a scaled image. Block averaging is
# done by a subprocedure; this procedure gets a line from a possibly block
# averaged image and if necessary interpolates it to the grid points of the
# output line.
pointer procedure sigm2r (si, lineno)
pointer si # pointer to SI descriptor
int lineno
pointer rawline, tempp, gp
int i, new_y[2], tempi, curbuf, altbuf
int nraw, npix, nblks_y, ybavg, x1, x2
real x, y, weight_1, weight_2
pointer si_blmavgr()
errchk si_blmavgr
begin
nraw = IM_LEN(SI_IM(si))
npix = SI_NPIX(si,1)
# Deterine the range of X (in pixels on the block averaged input image)
# required for the interpolator.
gp = SI_GRID(si,1)
x1 = SI_XOFF(si)
x = Memr[gp+npix-1]
x2 = x1 + int(x)
if (INTVAL(x))
x2 = x2 - 1
x2 = max (x1 + 1, x2)
gp = SI_GRID(si,2)
y = Memr[gp+lineno-1]
# The following is an optimization provided for the case when it is
# not necessary to interpolate in either X or Y. Block averaging is
# permitted.
if (SI_INTERP(si,1) == NO && SI_INTERP(si,2) == NO)
return (si_blmavgr (SI_IM(si), SI_FP(si), x1, x2, int(y),
SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si)))
# If we are interpolating in Y two buffers are required, one for each
# of the two input image lines required to interpolate in Y. The lines
# stored in these buffers are interpolated in X to the output grid but
# not in Y. Both buffers are not required if we are not interpolating
# in Y, but we use them anyhow to simplify the code.
if (SI_INIT(si) == YES) {
do i = 1, 2 {
if (SI_BUF(si,i) != NULL)
call mfree (SI_BUF(si,i), SI_TYBUF(si))
call malloc (SI_BUF(si,i), npix, TY_REAL)
SI_TYBUF(si) = TY_REAL
SI_BUFY(si,i) = NOTSET
}
if (OUTBUF(si) != NULL)
call mfree (OUTBUF(si), SI_TYBUF(si))
call malloc (OUTBUF(si), npix, TY_REAL)
SI_INIT(si) = NO
}
# If the Y value of the new line is not in range of the contents of the
# current line buffers, refill one or both buffers. To refill we must
# read a (possibly block averaged) input line and interpolate it onto
# the X grid. The X and Y values herein are in the coordinate system
# of the (possibly block averaged) input image.
new_y[1] = int(y)
new_y[2] = int(y) + 1
# Get the pair of lines whose integral Y values form an interval
# containing the fractional Y value of the output line. Sometimes the
# desired line will happen to be in the other buffer already, in which
# case we just have to swap buffers. Often the new line will be the
# current line, in which case nothing is done. This latter case occurs
# frequently when the magnification ratio is large.
curbuf = 1
altbuf = 2
do i = 1, 2 {
if (new_y[i] == SI_BUFY(si,i)) {
;
} else if (new_y[i] == SI_BUFY(si,altbuf)) {
SWAPP (SI_BUF(si,1), SI_BUF(si,2))
SWAPI (SI_BUFY(si,1), SI_BUFY(si,2))
} else {
# Get line and interpolate onto output grid. If interpolation
# is not required merely copy data out. This code is set up
# to always use two buffers; in effect, there is one buffer of
# look ahead, even when Y[i] is integral. This means that we
# will go out of bounds by one line at the top of the image.
# This is handled by copying the last line.
ybavg = SI_BAVG(si,2)
nblks_y = (IM_LEN (SI_IM(si), 2) + ybavg-1) / ybavg
if (new_y[i] <= nblks_y)
rawline = si_blmavgr (SI_IM(si), SI_FP(si), x1, x2,
new_y[i], SI_BAVG(si,1), SI_BAVG(si,2), SI_ORDER(si))
if (SI_INTERP(si,1) == NO) {
call amovr (Memr[rawline], Memr[SI_BUF(si,i)], npix)
} else if (SI_ORDER(si) == 0) {
call si_sampler (Memr[rawline], Memr[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
} else if (SI_ORDER(si) == -1) {
call si_maxr (Memr[rawline], nraw,
Memr[SI_GRID(si,1)], Memr[SI_BUF(si,i)], npix)
} else {
call aluir (Memr[rawline], Memr[SI_BUF(si,i)],
Memr[SI_GRID(si,1)], npix)
}
SI_BUFY(si,i) = new_y[i]
}
SWAPI (altbuf, curbuf)
}
# We now have two line buffers straddling the output Y value,
# interpolated to the X grid of the output line. To complete the
# bilinear interpolation operation we take a weighted sum of the two
# lines. If the range from SI_BUFY(si,1) to SI_BUFY(si,2) is repeatedly
# interpolated in Y no additional i/o occurs and the linear
# interpolation operation (ALUI) does not have to be repeated (only the
# weighted sum is required). If the distance of Y from one of the
# buffers is zero then we do not even have to take a weighted sum.
# This is not unusual because we may be called with a magnification
# of 1.0 in Y.
weight_1 = 1.0 - (y - SI_BUFY(si,1))
weight_2 = 1.0 - weight_1
if (weight_1 < SI_TOL)
return (SI_BUF(si,2))
else if (weight_2 < SI_TOL || SI_ORDER(si) == 0)
return (SI_BUF(si,1))
else if (SI_ORDER(si) == -1) {
call amaxr (Memr[SI_BUF(si,1)], Memr[SI_BUF(si,2)],
Memr[OUTBUF(si)], npix)
return (OUTBUF(si))
} else {
call awsur (Memr[SI_BUF(si,1)], Memr[SI_BUF(si,2)],
Memr[OUTBUF(si)], npix, weight_1, weight_2)
return (OUTBUF(si))
}
end
# SI_BLMAVGR -- Get a line from a block averaged image of type short.
# For example, block averaging by a factor of 2 means that pixels 1 and 2
# are averaged to produce the first output pixel, 3 and 4 are averaged to
# produce the second output pixel, and so on. If the length of an axis
# is not an integral multiple of the block size then the last pixel in the
# last block will be replicated to fill out the block; the average is still
# defined even if a block is not full.
pointer procedure si_blmavgr (im, fp, x1, x2, y, xbavg, ybavg, order)
pointer im # input image
pointer fp # fixpix structure
int x1, x2 # range of x blocks to be read
int y # y block to be read
int xbavg, ybavg # X and Y block averaging factors
int order # averaging option
int nblks_x, nblks_y, ncols, nlines, xoff, blk1, blk2, i, j, k
int first_line, nlines_in_sum, npix, nfull_blks, count
real sum, blkmax
pointer sp, a, b
pointer xt_fpr()
errchk xt_fpr
begin
call smark (sp)
ncols = IM_LEN(im,1)
nlines = IM_LEN(im,2)
xoff = (x1 - 1) * xbavg + 1
npix = min (ncols, xoff + (x2 - x1 + 1) * xbavg - 1) - xoff + 1
if ((xbavg < 1) || (ybavg < 1))
call error (1, "si_blmavg: illegal block size")
else if (x1 < 1 || x2 > ncols)
call error (2, "si_blmavg: column index out of bounds")
else if ((xbavg == 1) && (ybavg == 1))
return (xt_fpr (fp, im, y, NULL) + xoff - 1)
nblks_x = (npix + xbavg-1) / xbavg
nblks_y = (nlines + ybavg-1) / ybavg
if (y < 1 || y > nblks_y)
call error (2, "si_blmavg: block number out of range")
call salloc (b, nblks_x, TY_REAL)
if (ybavg > 1) {
call aclrr (Memr[b], nblks_x)
nlines_in_sum = 0
}
# Read and accumulate all input lines in the block.
first_line = (y - 1) * ybavg + 1
do i = first_line, min (nlines, first_line + ybavg - 1) {
# Get line from input image.
a = xt_fpr (fp, im, i, NULL) + xoff - 1
# Block average line in X.
if (xbavg > 1) {
# First block average only the full blocks.
nfull_blks = npix / xbavg
if (order == -1) {
blk1 = a
do j = 1, nfull_blks {
blk2 = blk1 + xbavg
blkmax = Memr[blk1]
do k = blk1+1, blk2-1
blkmax = max (blkmax, Memr[k])
Memr[a+j-1] = blkmax
blk1 = blk2
}
} else
call abavr (Memr[a], Memr[a], nfull_blks, xbavg)
# Now average the final partial block, if any.
if (nfull_blks < nblks_x) {
if (order == -1) {
blkmax = Memr[blk1]
do k = blk1+1, a+npix-1
blkmax = max (blkmax, Memr[k])
Memr[a+j-1] = blkmax
} else {
sum = 0.0
count = 0
do j = nfull_blks * xbavg + 1, npix {
sum = sum + Memr[a+j-1]
count = count + 1
}
Memr[a+nblks_x-1] = sum / count
}
}
}
# Add line into block sum. Keep track of number of lines in sum
# so that we can compute block average later.
if (ybavg > 1) {
if (order == -1)
call amaxr (Memr[a], Memr[b], Memr[b], nblks_x)
else {
call aaddr (Memr[a], Memr[b], Memr[b], nblks_x)
nlines_in_sum = nlines_in_sum + 1
}
}
}
# Compute the block average in Y from the sum of all lines block
# averaged in X. Overwrite buffer A, the buffer returned by IMIO.
# This is kosher because the block averaged line is never longer
# than an input line.
if (ybavg > 1) {
if (order == -1)
call amovr (Memr[b], Memr[a], nblks_x)
else
call adivkr (Memr[b], real(nlines_in_sum), Memr[a], nblks_x)
}
call sfree (sp)
return (a)
end
# SI_MAXR -- Resample a line via maximum value.
procedure si_maxr (a, na, x, b, nb)
real a[na] # input array
int na # input size
real x[nb] # sample grid
real b[nb] # output arrays
int nb # output size
int i
begin
do i = 1, nb
b[i] = max (a[int(x[i])], a[min(na,int(x[i]+1))])
end
|