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
|
/* pgmkernel.c - generate a PGM convolution kernel
**
** Creates a PGM image containing a convolution filter with max value = 255
** and minimum value > 127 that can be used as a smoothing kernel for
** pnmconvol.
**
** Copyright (C) 1992 by Alberto Accomazzi, Smithsonian Astrophysical
** Observatory.
**
** Permission to use, copy, modify, and distribute this software and its
** documentation for any purpose and without fee is hereby granted, provided
** that the above copyright notice appear in all copies and that both that
** copyright notice and this permission notice appear in supporting
** documentation. This software is provided "as is" without express or
** implied warranty.
*/
#include <math.h>
#include "pm_c_util.h"
#include "shhopt.h"
#include "mallocvar.h"
#include "pgm.h"
#include "nstring.h"
struct CmdlineInfo {
/* All the information the user supplied in the command line,
in a form easy for the program to use.
*/
unsigned int cols;
unsigned int rows;
float weight;
gray maxval;
};
static void
parseCommandLine(int argc, const char ** argv,
struct CmdlineInfo * const cmdlineP) {
/*----------------------------------------------------------------------------
Convert program invocation arguments (argc,argv) into a format the
program can use easily, struct cmdlineInfo. Validate arguments along
the way and exit program with message if invalid.
Note that some string information we return as *cmdlineP is in the storage
argv[] points to.
-----------------------------------------------------------------------------*/
optEntry *option_def;
/* Instructions to OptParseOptions2 on how to parse our options.
*/
optStruct3 opt;
unsigned int weightSpec, maxvalSpec;
unsigned int option_def_index;
MALLOCARRAY_NOFAIL(option_def, 100);
option_def_index = 0; /* incremented by OPTENTRY */
OPTENT3(0, "weight", OPT_FLOAT, &cmdlineP->weight,
&weightSpec, 0);
OPTENT3(0, "maxval", OPT_UINT, &cmdlineP->maxval,
&maxvalSpec, 0);
opt.opt_table = option_def;
opt.short_allowed = FALSE; /* We have no short (old-fashioned) options */
opt.allowNegNum = FALSE; /* We have no parms that are negative numbers */
pm_optParseOptions3(&argc, (char **)argv, opt, sizeof(opt), 0);
/* Uses and sets argc, argv, and some of *cmdlineP and others. */
if (!weightSpec)
cmdlineP->weight = 6.0;
if (cmdlineP->weight < 0.0)
pm_error("-weight cannot be negative. You specified %f",
cmdlineP->weight);
if (!maxvalSpec)
cmdlineP->maxval = PGM_MAXMAXVAL;
if (cmdlineP->maxval > PGM_OVERALLMAXVAL)
pm_error("-maxval is too large: %u. Maximum is %u",
cmdlineP->maxval, PGM_OVERALLMAXVAL);
if (cmdlineP->maxval == 0)
pm_error("-maxval cannot be zero");
if (argc-1 < 1)
pm_error("Need at least one argument: size of (square) kernel");
else if (argc-1 == 1) {
unsigned int dimension;
const char * error;
pm_string_to_uint(argv[1], &dimension, &error);
if (error) {
pm_error("'%s' is invalid as an image width/height. %s", argv[1], error);
pm_strfree(error);
}
if (dimension <= 0)
pm_error("Dimension must be a positive number. "
"You specified '%s'", argv[1]);
cmdlineP->cols = cmdlineP->rows = dimension;
} else if (argc-1 == 2) {
unsigned int const width = pm_parse_width(argv[1]);
unsigned int const height = pm_parse_height(argv[2]);
if (width <= 0)
pm_error("Width must be a positive number. "
"You specified '%s'", argv[1]);
if (height <= 0)
pm_error("Height must be a positive number. "
"You specified '%s'", argv[2]);
cmdlineP->cols = width;
cmdlineP->rows = height;
} else
pm_error("At most two arguments allowed. "
"You specified %u", argc-1);
}
static double
t(double const dx2,
double const dy2,
double const weight) {
/*----------------------------------------------------------------------------
The t value for a pixel that is (dx, dy) pixels away from the center of
the kernel, where 'dx2' is SQR(dx) and 'dy2' is SQR(dy), if the distance is
weighted by 'weight'.
-----------------------------------------------------------------------------*/
return 1.0 / (1.0 + weight * sqrt(dx2 + dy2));
}
static double
tMaxAllKernel(unsigned int const cols,
unsigned int const rows,
double const weight) {
/*----------------------------------------------------------------------------
The maximum t value over all pixels in the kernel, if the kernel is
'cols' by 'rows' pixels and distance is weighted by 'weight'.
-----------------------------------------------------------------------------*/
/* It depends upon whether there is an even or odd number of rows
and columns. If both dimensions are odd, there is a pixel right
at the center, and it has the greatest t value. If both dimensions
are even, the center of the image is in the center of a 4-pixel
square and each of those 4 pixels has the greatest t value. If
one dimension is even and the other odd, the center of the kernel
is midway between two pixels, horizontally or vertically, and one
of those two pixels has the greatest t value.
*/
double dxMax, dyMax;
switch (cols % 2 + rows % 2) {
case 0:
dxMax = 0.5;
dyMax = 0.5;
break;
case 1:
dxMax = 0.5;
dyMax = 0.0;
break;
case 2:
dxMax = 0.0;
dyMax = 0.0;
}
return t(SQR(dxMax), SQR(dyMax), weight);
}
static void
writeKernel(FILE * const ofP,
unsigned int const cols,
unsigned int const rows,
gray const maxval,
gray ** const halfKernel,
unsigned int const halfRows) {
unsigned int row;
pgm_writepgminit(stdout, cols, rows, maxval, 0);
for (row = 0; row < halfRows; ++row)
pgm_writepgmrow(stdout, halfKernel[row], cols, maxval, 0);
/* Now write out the same rows in reverse order. */
for (; row < rows; ++row)
pgm_writepgmrow(stdout, halfKernel[rows-1-row], cols, maxval, 0);
}
int
main(int argc, const char * argv[]) {
struct CmdlineInfo cmdline;
unsigned int arows;
unsigned int arow;
double xcenter, ycenter;
/* row, column "number" of center of kernel */
double tMax;
/* The maximum t value over all pixels */
gray ** halfKernel;
/* The upper half of the kernel we generate. The lower half is
just the mirror image of this.
*/
pm_proginit(&argc, argv);
parseCommandLine(argc, argv, &cmdline);
xcenter = ((double) cmdline.cols - 1) / 2.0;
ycenter = ((double) cmdline.rows - 1) / 2.0;
tMax = tMaxAllKernel(cmdline.cols, cmdline.rows, cmdline.weight);
/* Output matrix is symmetric vertically and horizontally. */
arows = (cmdline.rows + 1) / 2;
/* Half the number of rows. Add 1 if odd. */
halfKernel = pgm_allocarray(cmdline.cols, arows);
for (arow = 0; arow < arows; ++arow) {
double const dy2 = SQR(arow - ycenter);
unsigned int col;
for (col = 0; col < (cmdline.cols +1) / 2; ++col) {
double const epsilon = 1e-15;
double const dx2 = SQR(col - xcenter);
double const normalized = t(dx2, dy2, cmdline.weight) / 2 / tMax;
gray const grayval =
ROUNDU(cmdline.maxval * (0.5 + normalized + epsilon));
halfKernel[arow][col ] = grayval;
halfKernel[arow][cmdline.cols - col - 1] = grayval;
}
}
writeKernel(stdout, cmdline.cols, cmdline.rows, cmdline.maxval,
halfKernel, arows);
pgm_freearray(halfKernel, arows);
return 0;
}
|