| 11 |
|
][ |
| 12 |
|
.B \-f[afdc] |
| 13 |
|
][ |
| 14 |
+ |
.B "\-n nproc" |
| 15 |
+ |
][ |
| 16 |
|
.B "\-f file" |
| 17 |
|
][ |
| 18 |
|
.B "\-e expr" |
| 55 |
|
.I \-s |
| 56 |
|
options follow the last input matrix, output results will be transformed |
| 57 |
|
and/or scaled accordingly. |
| 58 |
< |
These operations are discussed in greater detail further on. |
| 58 |
> |
These operations are discussed in greater detail below. |
| 59 |
|
A single concatenation matrix may be applied after element operations |
| 60 |
|
using the |
| 61 |
|
.I \-m |
| 73 |
|
.fi |
| 74 |
|
.sp |
| 75 |
|
The number of components indicates that each matrix element is actually |
| 76 |
< |
composed of multiple elements, most commonly an RGB triple. |
| 76 |
> |
composed of multiple channels, most commonly an RGB triple. |
| 77 |
|
This is essentially dividing the matrix into planes, where each component |
| 78 |
|
participates in a separate calculation. |
| 79 |
|
If an appropriate header is not present, it may be added with a call to |
| 96 |
|
option. |
| 97 |
|
Radiance spectral pictures with more than 3 components are also supported. |
| 98 |
|
These are typically produced by |
| 99 |
< |
.I rtrace(1) |
| 99 |
> |
.I rtpict(1) |
| 100 |
|
or |
| 101 |
|
.I rfluxmtx(1). |
| 102 |
|
.PP |
| 143 |
|
is specified for an input picture or the |
| 144 |
|
.I "-fc" |
| 145 |
|
option is given, the output will be written as a RGBE or XYZE picture. |
| 146 |
< |
Note that conversion from a float or RGBE color space applies a conversion |
| 146 |
> |
Note that conversion from a float or RGBE color space applies an efficacy factor |
| 147 |
|
of 179 lumens/watt (for CIE or melanopic output) or 412 (for scotopic output), |
| 148 |
< |
and the reverse happens for conversion from XYZE input to RGB or RGBE output. |
| 148 |
> |
and the inverse happens for conversion from XYZE input to RGB or RGBE output. |
| 149 |
> |
Lower case versions of all these components are also supported, the only |
| 150 |
> |
difference being that the efficacy factors are ignored. |
| 151 |
|
.PP |
| 152 |
|
If a matrix or picture file path is given to the |
| 153 |
|
.I \-c |
| 189 |
|
or |
| 190 |
|
.I \-f |
| 191 |
|
options are used to define a "co" variable or "co(p)" function, |
| 192 |
< |
this will be evaluated at each output |
| 193 |
< |
component for the current element. |
| 192 |
> |
which will be evaluated for each output |
| 193 |
> |
component from the current element. |
| 194 |
|
The "co" variable defines identical operations for all components, |
| 195 |
|
whereas "co(p)" may specify different operations for each component. |
| 196 |
|
The element position is defined |
| 199 |
|
goes from 0 to "nrows" minus one, and |
| 200 |
|
.I c |
| 201 |
|
goes from 0 to "ncols" minus one. |
| 202 |
+ |
(Note that "nrows" may be zero if unspecified in inputs, and this |
| 203 |
+ |
is a unique capability of |
| 204 |
+ |
.I rcomb |
| 205 |
+ |
to handle these.)\0 |
| 206 |
|
Component p from input i is accessed with the "ci(i,p)" function, |
| 207 |
|
and the number of components is defined by the "ncomp" constant. |
| 208 |
|
If given as "ci(i)", the function returns the current component |
| 209 |
|
being evaluated by |
| 210 |
|
.I rcomb. |
| 211 |
< |
A different component may be referenced using th second argument. |
| 211 |
> |
A different component may be referenced using the second argument. |
| 212 |
|
For example, "ci(1,2)" accesses |
| 213 |
|
the second component from the first input. |
| 214 |
|
If the input is a picture, the the constants "R", "G", and "B" |
| 223 |
|
the functions "ri(i)", "gi(i)", and "bi(i)" are predefined as |
| 224 |
|
"ci(i,R)", "ci(i,G)", and "ci(i,B)", respectively. |
| 225 |
|
Accordingly, the "ro", "go", and "bo" |
| 226 |
< |
variables may be used in place of "co(R)", "co(G)", and "co(B)". |
| 226 |
> |
variables may be used in place of "co(R)", "co(G)", and "co(B)", |
| 227 |
> |
but all three must be defined for this substitution to take place. |
| 228 |
|
Finally, the total number of input files is set in the constant "nfiles". |
| 229 |
|
.PP |
| 230 |
|
Results are sent to the standard output. |
| 251 |
|
.I \-w |
| 252 |
|
option turns off warnings about divide-by-zero and other non-fatal |
| 253 |
|
calculation errors. |
| 254 |
+ |
.PP |
| 255 |
+ |
The |
| 256 |
+ |
.I \-n |
| 257 |
+ |
option specifies how many execution processes to employ, |
| 258 |
+ |
which may improve performance on multi-core architectures, |
| 259 |
+ |
especially for matrix multiplication |
| 260 |
+ |
and complex operations on long input rows. |
| 261 |
|
.SH EXAMPLES |
| 262 |
< |
To convert two hyperspectral inputs to RGB color space, |
| 262 |
> |
To convert two hyperspectral pictures to RGB color space, |
| 263 |
|
average them together, and write them out as a RADIANCE picture: |
| 264 |
|
.IP "" .2i |
| 265 |
< |
rcomb -C RGB -s .5 img1.spc -s .5 img2.spc > avg.hdr |
| 265 |
> |
rcomb -C RGB -s .5 img1.hsr -s .5 -fc img2.hsr > avg.hdr |
| 266 |
|
.PP |
| 267 |
|
Divide one set of matrix elements by the Euclidean sum of two others: |
| 268 |
|
.IP "" .2i |
| 272 |
|
Compute the absolute and relative differences between melanopic and photopic values |
| 273 |
|
in a spectral image: |
| 274 |
|
.IP "" .2i |
| 275 |
< |
rcomb -fa -C MY -e "abs(x):if(x,x,-x)" |
| 275 |
> |
rcomb -C MY -e "abs(x):if(x,x,-x)" |
| 276 |
|
-e "co(p)=select(p,abs(ci(1,1)-ci(1,2)),(ci(1,1)-ci(1,2))/ci(1,2))" |
| 277 |
< |
input_spec.hsr > compare.mtx |
| 277 |
> |
input_spec.hsr -fa > compare.mtx |
| 278 |
|
.PP |
| 279 |
|
Concatenate a spectral flux coefficient matrix with a spectral sky |
| 280 |
|
matrix to compute a set of melanopic lux values: |
| 289 |
|
.I pcomb, |
| 290 |
|
whose capabilities somewhat overlap. |
| 291 |
|
The former loads each matrix into memory before operations, |
| 292 |
< |
and element components take 8 bytes apiece, adding up quickly. |
| 292 |
> |
and element components are stored as double-precision. |
| 293 |
|
Very large matrices therefore present a problem with that tool. |
| 294 |
|
Furthermore, |
| 295 |
|
.I rmtxop |
| 322 |
|
.I rcomb |
| 323 |
|
should be preferred over |
| 324 |
|
.I rmtxop |
| 325 |
< |
for any operations in can handle, which is everything except |
| 325 |
> |
for any operations it can handle, which is everything except |
| 326 |
|
multiple matrix concatenations and transpose |
| 327 |
< |
operations, which are handled more efficiently by |
| 328 |
< |
.I rcollate(1) |
| 329 |
< |
in any case. |
| 327 |
> |
operations. |
| 328 |
> |
The latter may be handled more efficiently by |
| 329 |
> |
.I rcollate(1). |
| 330 |
|
That said, there is no significant difference for |
| 331 |
< |
simple operations on smallish matrices, and note that only |
| 331 |
> |
simple operations on small matrices, and only |
| 332 |
|
.I rmtxop |
| 333 |
|
and |
| 334 |
|
.I dctimestep(1) |
| 335 |
< |
currently accept XML files as inputs. |
| 336 |
< |
Also, the resizing function of |
| 335 |
> |
accept XML files as inputs. |
| 336 |
> |
Also note that the resizing function of |
| 337 |
|
.I pcomb |
| 338 |
|
is not supported in |
| 339 |
|
.I rcomb, |
| 340 |
|
and should instead be handled by |
| 341 |
|
.I pfilt(1). |
| 342 |
+ |
.SH BUGS |
| 343 |
+ |
The |
| 344 |
+ |
.I rcomb |
| 345 |
+ |
command currently ignores the "PRIMARIES" setting in input |
| 346 |
+ |
headers, and does not produce any on output, even in |
| 347 |
+ |
circumstances where it would make sense to. |
| 348 |
|
.SH AUTHOR |
| 349 |
|
Greg Ward |
| 350 |
|
.SH "SEE ALSO" |
