1 | Duchamp version 1.3.2 (2013/05/02) |
---|
2 | ---------------------------------- |
---|
3 | |
---|
4 | Bug fixes: |
---|
5 | |
---|
6 | * The units for the integrated flux (for F_int) were not being |
---|
7 | calculated for cases of a FITS file without a spectral |
---|
8 | axis. Additionally, a vestigial default value meant this column had |
---|
9 | units of "counts". This has been fixed so that a FITS file with |
---|
10 | Jy/beam units will have F_int values in Jy. |
---|
11 | |
---|
12 | * The ellipse fitting to find the MAJ/MIN/PA was not working correctly |
---|
13 | in the case of negative searches. This has been fixed. |
---|
14 | |
---|
15 | * The scaling of MAJ/MIN values was not being done correctly in all |
---|
16 | cases, so that the values were out by a factor of 60 or 3600. This |
---|
17 | was particularly the case when the ellipse fitting failed (due to |
---|
18 | small numbers of pixels, for instance). |
---|
19 | |
---|
20 | * The principle axes calculations have been simplified slightly, and |
---|
21 | now are done only on pixels above half the peak in the moment-0 map |
---|
22 | (to estimate the FWHM). |
---|
23 | |
---|
24 | Duchamp version 1.3.1 (2013/04/18) |
---|
25 | ---------------------------------- |
---|
26 | |
---|
27 | Bug fixes: |
---|
28 | |
---|
29 | * The kernel used in the 2D smoothing was not normalised properly, |
---|
30 | which meant the fluxes in the smoothed array were scaled |
---|
31 | incorrectly. The kernel is now normalised, so that the fluxes in the |
---|
32 | smoothed array decrease (and to an increasing degree with larger |
---|
33 | kernels, as expected). |
---|
34 | |
---|
35 | * The way the position angle of the smoothing kernel was interpreted |
---|
36 | was wrong. It now conforms to the typical astronomical position |
---|
37 | angle definition of degrees east of north. |
---|
38 | |
---|
39 | * The Z_PEAK column was being left out of the results file for |
---|
40 | Duchamp-1.3. It has been put back in. |
---|
41 | |
---|
42 | * The calculation of S/Nmax has changed so that when smoothing or |
---|
43 | reconstruction is done, the peak flux used is measured in the |
---|
44 | reconstructed/smoothed array (where the noise is measured). |
---|
45 | |
---|
46 | * The flux errors are now not reported in the case of smoothing, since |
---|
47 | we don't measure the noise in the original input image array. |
---|
48 | |
---|
49 | Other fixes: |
---|
50 | |
---|
51 | * The vertical range of the spectral plots now properly takes into |
---|
52 | account all spectra (not just the input array), and quirks with a MW |
---|
53 | range lying outside the plotted range have been fixed. |
---|
54 | |
---|
55 | * The flux scaling for the moment-0 map plot now ignores negative |
---|
56 | values, meaning the scale should be sensible. |
---|
57 | |
---|
58 | * The spelling of the Object2D::getPrincipalAxes() has been corrected. |
---|
59 | |
---|
60 | * Minor code optimisations have been made to the smoothing code. |
---|
61 | |
---|
62 | * The subsection class is now initialised with a blank string. |
---|
63 | |
---|
64 | Duchamp version 1.3 (2013/03/08) |
---|
65 | -------------------------------- |
---|
66 | |
---|
67 | New features: |
---|
68 | |
---|
69 | * New source parameters: |
---|
70 | |
---|
71 | # Each source now has a measurement of the major & minor axis sizes, |
---|
72 | along with the position angle of the major axis. The sizes are the |
---|
73 | full-width at half maximum, measured from the moment-0 map. These |
---|
74 | now appear in the catalogue and in the spectral plots, with the |
---|
75 | ellipse drawn on the moment map cutout in the spectral plots. The |
---|
76 | old measurements of w_RA & w_DEC are still made and included in |
---|
77 | the full catalogue, but are no longer reported in the smaller |
---|
78 | catalogue printed to screen (nor in the XML/VOTable output). |
---|
79 | |
---|
80 | # Each source also now has a measurement of the error on the |
---|
81 | integrated and total fluxes. This is calculated from summing in |
---|
82 | quadrature the statistical noise on each detected voxel. This is |
---|
83 | from a request by Sara Shakouri. |
---|
84 | |
---|
85 | * Output files - new types and changes to the way they are written: |
---|
86 | |
---|
87 | # Ability to create CASA region format files showing the location of |
---|
88 | the detected sources. This is analogous to the Karma annotation |
---|
89 | files or DS9 region files, for use with the casa viewer, but can |
---|
90 | also be imported into casapy. |
---|
91 | |
---|
92 | # Ability to save baseline values to a FITS file (in the same way as |
---|
93 | for the reconstructed or smoothed array). |
---|
94 | |
---|
95 | # Ability to save the moment-0 mask to a FITS file (that is, a mask |
---|
96 | indicating which spatial pixels form part of the moment-0 map). |
---|
97 | |
---|
98 | # There is an improved way to store the detected pixels for later |
---|
99 | re-use. While previously the detected pixels were written as scans |
---|
100 | to the logfile, and then read in when the usePrevious flag was |
---|
101 | set, we now write the scans and other metadata to a binary |
---|
102 | catalogue that can be read in again. Writing as a binary file |
---|
103 | leads to much better compression of the data, particularly when |
---|
104 | combined with the run-length encoding. It is anticipated future |
---|
105 | tools will be provided to allow additional interaction with these |
---|
106 | binary catalogues. |
---|
107 | |
---|
108 | # There are slight changes to the format of the ASCII catalogues, |
---|
109 | where lines not containing sources are prefixed with a comment |
---|
110 | character '#'. This will facilitate their use in other software. |
---|
111 | |
---|
112 | * There is a new command-line parameter, -t THRESH, which specifies |
---|
113 | the flux threshold for searching. It is designed to be used with the |
---|
114 | -f option, as a quick way of searching a FITS image to a specified |
---|
115 | flux level. |
---|
116 | |
---|
117 | Bug fixes: |
---|
118 | |
---|
119 | * A bug in the spatial smoothing has been fixed. The mask was being |
---|
120 | interpreted incorrectly, so that no valid pixels were getting |
---|
121 | smoothed. |
---|
122 | |
---|
123 | * The WCS for the moment-0 FITS cube has been fixed (specifically, the |
---|
124 | axis names). |
---|
125 | |
---|
126 | * When usePrevious was used, the behaviour in previous versions was to |
---|
127 | always write individual spectra. This is now controlled by a new |
---|
128 | parameter flagPlotIndividualSpectra. |
---|
129 | |
---|
130 | * Minor improvements to the output when verbose=false. |
---|
131 | |
---|
132 | * Improvements to the Makefile, so that header files from a |
---|
133 | previously-installed version of Duchamp do not get used instead of |
---|
134 | those from the current version. |
---|
135 | |
---|
136 | * The vertical range for the spectral plots now take into account the |
---|
137 | baseline and reconstructed/smoothed arrays as well (outside of any |
---|
138 | MW range, of course). |
---|
139 | |
---|
140 | Other |
---|
141 | |
---|
142 | * The code-base for reading from and writing to FITS files has been |
---|
143 | overhauled and streamlined. It can now be found in the FitsIO |
---|
144 | directory. The interface for the writing of FITS files via the Cube |
---|
145 | class has not changed. |
---|
146 | |
---|
147 | * The User Guide has been slightly re-ordered to provide more explicit |
---|
148 | information about the source parameters. |
---|
149 | |
---|
150 | |
---|
151 | Duchamp version 1.2.2 (2012/10/29) |
---|
152 | ---------------------------------- |
---|
153 | |
---|
154 | Bug fixes: |
---|
155 | |
---|
156 | * The reading of reconstructed & smoothed arrays in the case of large |
---|
157 | (>2GB) cubes has been improved. While the most recent versions of |
---|
158 | cfitsio were able to handle these, earlier versions would result in |
---|
159 | a segementation fault using the previous interface. An alternative |
---|
160 | interface has been implemented that avoids these faults, even for |
---|
161 | the older cfitsio libraries. |
---|
162 | |
---|
163 | * A bug affecting the writing of FITS mask files (apparently for the |
---|
164 | case of flagging by the object number) has been fixed. |
---|
165 | |
---|
166 | * The moment map occasionally failed if sources lay spatially within |
---|
167 | the blank pixel region of the first channel. This check has been |
---|
168 | removed and the moment map should be produced successfully in all |
---|
169 | situations. |
---|
170 | |
---|
171 | Other: |
---|
172 | |
---|
173 | * The UCDs used in the VOTable catalogue output have been |
---|
174 | updated. Several of these were shared between different columns, and |
---|
175 | so they have been changed so that each UCD is unique to one |
---|
176 | column. The column ID strings have been updated for clarity. |
---|
177 | |
---|
178 | * The determination of default output image filenames has been |
---|
179 | improved -- it previously assumed the input filename ended in |
---|
180 | '.fits', but this is no longer the case. |
---|
181 | |
---|
182 | Duchamp version 1.2.1 (2012/10/09) |
---|
183 | ---------------------------------- |
---|
184 | |
---|
185 | New feature: |
---|
186 | |
---|
187 | * Duchamp can now produce region files to annotate images displayed in |
---|
188 | SAOImage DS9. There are new parameters flagDS9 and ds9File that work |
---|
189 | in the same way as the Karma annotation file parameters. The region |
---|
190 | files can also be displayed in (at least some versions of) |
---|
191 | casaviewer. |
---|
192 | |
---|
193 | * There is a new script to fix occasional problems with the |
---|
194 | automatically-detected Makefile, and the verification script has |
---|
195 | been updated and improved. |
---|
196 | |
---|
197 | Bug fixes: |
---|
198 | |
---|
199 | * The main bug fix for this release corrects a problem with the |
---|
200 | velocity width measurements made when searching for negative |
---|
201 | detections. It was possible to get very small (and incorrect) widths |
---|
202 | in some cases - this was due to the integrated spectrum not being |
---|
203 | inverted correctly. |
---|
204 | |
---|
205 | * In the case of smoothing, the stats of the full cube that were being |
---|
206 | reported in the output files were actually those of the un-smoothed |
---|
207 | (input cube). This only affected the output files, not the detection |
---|
208 | process itself. |
---|
209 | |
---|
210 | * The units were not being reported in the VOTable. This has been |
---|
211 | fixed. |
---|
212 | |
---|
213 | * The UCDs for the spectral parameters in the VOTable were not being |
---|
214 | adapted to the user requests, so didn't always match the types of |
---|
215 | the parameters. All should be consistent now. |
---|
216 | |
---|
217 | Other changes: |
---|
218 | |
---|
219 | * The codebase relating to the creation of and writing to output files |
---|
220 | (results, logs, annotation, VOTables) has been overhauled, with a |
---|
221 | cleaner interface. Most of this code is in the new subdirectory |
---|
222 | src/Outputs, to where the columns code has been moved as well. |
---|
223 | |
---|
224 | This was instigated by ASKAP/Selavy development, but also allowed |
---|
225 | the expansion of output formats to include DS9 region files. No |
---|
226 | change to the user should be visible, but anyone using the software |
---|
227 | will notice differences. Please get in touch if changes are |
---|
228 | difficult to follow. |
---|
229 | |
---|
230 | Duchamp version 1.2 (2012/08/08) |
---|
231 | -------------------------------- |
---|
232 | |
---|
233 | Key changes: |
---|
234 | |
---|
235 | * The way the wavelet reconstruction is done in the presence of BLANK |
---|
236 | voxels is slightly different, specifically when there were BLANK |
---|
237 | voxels on the edge. While BLANK voxels are always ignored by the |
---|
238 | reconstruction (and left in place), the previous approach dealt with |
---|
239 | BLANKS on the edge by moving the effective edge of the image inwards |
---|
240 | until the first non-blank voxel. This meant that the size of the |
---|
241 | dimensions used to determine the number of scales, and the location |
---|
242 | where boundary reflections took place where different to the input |
---|
243 | image size. |
---|
244 | |
---|
245 | However, this meant that BLANK voxels on the edge were treated |
---|
246 | differently to those not on the edge, so this inconsistency was |
---|
247 | removed. The full image dimensions are always used to calculate the |
---|
248 | number of scales, and reflections (which is how the reconstruction |
---|
249 | deals with the boundaries) will occur at the actual image edges. |
---|
250 | |
---|
251 | If your data has this property (BLANK pixels at the image |
---|
252 | boundaries), then any reconstruction will be affected (most likely |
---|
253 | only in a small way) and potentially any subsequent searching. You |
---|
254 | are advised to check this (for instance by saving the results of the |
---|
255 | resonstruction from 1.1.13 and 1.2 and comparing). |
---|
256 | |
---|
257 | * The way the spectral world coordinate system is dealt with has been |
---|
258 | improved. Previously Duchamp tried to convert to a velocity axis, |
---|
259 | and got a bit grumpy if that didn't work well. Now it will express |
---|
260 | the spectral units in the axis type indicated by the FITS |
---|
261 | header. Duchamp will also allow you to change this type by providing |
---|
262 | an appropriate WCS type as an input parameter. You may also provide |
---|
263 | a rest frequency in the parameter file and this will be used to |
---|
264 | calculate velocities. And, as before, you can provide specific units |
---|
265 | for the spectral quantities to be converted into. |
---|
266 | |
---|
267 | * Since v1.1.13 was released, the journal paper describing Duchamp has |
---|
268 | been published. It can be found at MNRAS 421, 3242 (2012), or |
---|
269 | http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2966.2012.20548.x/full. It |
---|
270 | includes descriptions of the various algorithms used for Duchamp, as |
---|
271 | well as some indicative comparisons of the relative completeness & |
---|
272 | reliability. Please cite this paper when describing Duchamp. |
---|
273 | |
---|
274 | Other new features: |
---|
275 | |
---|
276 | * The default value of snrCut has changed from 3 to 5. This is |
---|
277 | espeically important if you use the -f option, but also if you don't |
---|
278 | provide snrCut in your parameter file. |
---|
279 | |
---|
280 | * The XML VOTable output now includes a more complete list of |
---|
281 | parameters, matching the output given in text form in the results & |
---|
282 | log output files. Only those parameters affecting the results are |
---|
283 | included - things like the output file names and so on are left out. |
---|
284 | |
---|
285 | * The minVoxels parameter is now adjusted to be at least (minPix + |
---|
286 | minChannels - 1). If it is changed, a warning message is triggered. |
---|
287 | |
---|
288 | * The warning/error reporting has been overhauled, with an improved |
---|
289 | interface for developers. Also, some ERROR instances that previously |
---|
290 | resulted in the program terminating now allow it to continue |
---|
291 | (provided that is possible given the error). |
---|
292 | |
---|
293 | * It is now possible to give the convergence criterion for the wavelet |
---|
294 | reconstruction via the parameter reconConvergence. When the relative |
---|
295 | change in standard deviation of the residuals from the |
---|
296 | reconstruction falls below this value, the a trous algorithm stops |
---|
297 | iterating. |
---|
298 | |
---|
299 | * The way one-dimensional arrays were dealt with was unsatisfactory, |
---|
300 | both with the way positions were registered and the format of the |
---|
301 | graphical output. Improvements have been made such that a 1D data |
---|
302 | structure is assumed to be a spectrum, and so positions in the |
---|
303 | spectrum are reported as z-pixels (or the equivalent world |
---|
304 | coordinate). Instead of a 2D moment map, the X-windows display will |
---|
305 | show the spectrum with all detected pixels plus the final list of |
---|
306 | detections highlighted, along with the detection and growth |
---|
307 | thresholds. This plot will also be written to the "detectionMap" |
---|
308 | image. |
---|
309 | |
---|
310 | * All spectral plots will now show the detection (and growth, if |
---|
311 | applicable) thresholds when plotting the peak flux spectrum. |
---|
312 | |
---|
313 | * Accessing FITS files when a subsection string is provided has been |
---|
314 | streamlined. The cfitsio package, which handles the FITS access, |
---|
315 | creates a temporary file containing this image section, and this |
---|
316 | meant the previous approach was slow, particularly for large |
---|
317 | files/sections. Now the file is only opened once and the overhead of |
---|
318 | creating the temporary file is only encountered once. |
---|
319 | |
---|
320 | * Several aspects of the processing & output did not handle a |
---|
321 | non-existent WCS well. This has been improved, so that all stages of |
---|
322 | execution will proceed without errors. |
---|
323 | |
---|
324 | Bug fixes: |
---|
325 | |
---|
326 | * The absence of a BLANK keyword when trimming will now not trigger a |
---|
327 | failure, only a warning message. |
---|
328 | |
---|
329 | * The flux unit conversion (eg. from Jy/beam to mJy/beam) was not |
---|
330 | being applied correctly when an existing reconstructed or smoothed |
---|
331 | array was being read in. This is now applied properly. |
---|
332 | |
---|
333 | * When saving reconstructed or smoothed arrays, the data type is now |
---|
334 | set to FLOAT, so that the correct values are written (and |
---|
335 | subsequently read in). The previous version was potentially saving |
---|
336 | to a FITS file with integer data type (this depended on the type of |
---|
337 | the input array). In this case, the full precision of data values |
---|
338 | would not be preserved, so reading back in and doing an identical |
---|
339 | search could yield different sources. |
---|
340 | |
---|
341 | * For certain cases, the reconstruction attempted to use a very large |
---|
342 | number of scales. This has been fixed. |
---|
343 | |
---|
344 | * The maximum scale for the reconstruction was not always being |
---|
345 | applied. The scale ranges are now enforced for all types of |
---|
346 | reconstructions. |
---|
347 | |
---|
348 | * When saving a mask image while using a subsection of a cube, the |
---|
349 | offsets for the subsection were not being accounted for |
---|
350 | correctly. This is now fixed. |
---|
351 | |
---|
352 | * The detection map plots were not recording pixels added from the |
---|
353 | growing stage. This has been fixed. |
---|
354 | |
---|
355 | * The RA & Dec now have appropriate precision, rather than an |
---|
356 | arbitrary 2 decimal places on the seconds value. The RA will be one |
---|
357 | decimal place more than the Dec, and they will reflect the degree of |
---|
358 | precision in the image pixels. |
---|
359 | |
---|
360 | * A number of data type conversions (typically ints to floats) led to |
---|
361 | errors in some calculations, such as WCS conversions. Care has been |
---|
362 | taken to make sure all indexing is done appropriately with no |
---|
363 | overflow, and that all conversions (where necessary) are done |
---|
364 | appropriately. |
---|
365 | |
---|
366 | * It was possible to get world positions looking like 11:59:60.0! Not |
---|
367 | any more. |
---|
368 | |
---|
369 | * Reading previous detections from the log file (via the use of the |
---|
370 | usePrevious parameter) did result in error messages due to slightly |
---|
371 | incompatible formats. These have been fixed, and all graphical |
---|
372 | output should work properly. |
---|
373 | |
---|
374 | * When using a spectral search, having a blank pixel in the first |
---|
375 | channel could lead to that entire spectrum being ignored, leading to |
---|
376 | inconsistent results between spectral and spatial searches. This |
---|
377 | issue has been fixed, with both types of searches returning the same |
---|
378 | catalogue. |
---|
379 | |
---|
380 | |
---|
381 | Code base development notes: |
---|
382 | |
---|
383 | * Most instances of integer overflow due to insufficient precision |
---|
384 | have been tracked down - many of these were indexing variables of |
---|
385 | type int, which have been changed to size_t. |
---|
386 | |
---|
387 | * Importantly, for code interfacing with the duchamp::Cube and related |
---|
388 | classes, axisDim is now an array of size_t elements, rather than |
---|
389 | long. |
---|
390 | |
---|
391 | * Object3D now carries its spatial map with it, rather than |
---|
392 | calculating it on the fly when requested. This saves a lot of |
---|
393 | unnecessary computational effort. |
---|
394 | |
---|
395 | * The warning/error interface has improved, making use of C++ |
---|
396 | exception handling. This should be compatible with external logging |
---|
397 | infrastructure (such as log4cxx). |
---|
398 | |
---|
399 | * The interface to the FitsHeader & WCS structures for external code |
---|
400 | has been improved, so that the WCS can be set up via function calls. |
---|
401 | |
---|
402 | |
---|
403 | |
---|
404 | Duchamp version 1.1.13 (2011/04/10) |
---|
405 | ------------------------------------- |
---|
406 | |
---|
407 | Bug fixes: |
---|
408 | |
---|
409 | * The median statistics calculations had a bug in the test for whether |
---|
410 | the array under consideration was of even size. The correction for |
---|
411 | this can mean small differences in the resulting values. This can |
---|
412 | flow through to calculated detection thresholds - please check this |
---|
413 | with your data if you use this mode! |
---|
414 | |
---|
415 | * The reconstruction code had a bug where the incorrect number of |
---|
416 | scales was being used. This occurred when the blank pixel flag was |
---|
417 | set, and when the spectral dimension was smaller than both the |
---|
418 | spatial dimensions. |
---|
419 | |
---|
420 | Other: |
---|
421 | |
---|
422 | * The default value of the blank pixel flag has been set to false, |
---|
423 | rather than true. For Duchamp, this should not matter, as it is set |
---|
424 | when the FITS file is read in. For external code using, for |
---|
425 | instance, the reconstruction algorithms, this default may be |
---|
426 | important. Having it true was contributing to the second bug listed |
---|
427 | above. It also makes more sense for it to be false unless explicitly |
---|
428 | set in some way. |
---|
429 | |
---|
430 | * The reconstruction code has been streamlined a little, reducing its |
---|
431 | memory impact and using more appropriate variable types |
---|
432 | (eg. unsigned ints instead of ints and so forth). |
---|
433 | |
---|
434 | * For those who care about the actual code base, the statistics |
---|
435 | functions have been expanded, and separated from getStats.cc into |
---|
436 | "normal" and "robust" versions. |
---|
437 | |
---|
438 | |
---|
439 | Duchamp version 1.1.12 (2011/03/18) |
---|
440 | ------------------------------------- |
---|
441 | |
---|
442 | Bug fixes: |
---|
443 | |
---|
444 | * The velocity width calculations had a bug that occasionally led to |
---|
445 | very large values. The widths are calculated using the integrated |
---|
446 | spectrum, using a threshold relative to its peak. In v.1.1.10 and |
---|
447 | 1.1.11, however, the peak was searched for over the entire spectral |
---|
448 | range, rather than just the range for the current object. This led |
---|
449 | to spurious values. This has been fixed, so that the peak is only |
---|
450 | searched for within the detected ranage of the object. |
---|
451 | |
---|
452 | * Both the pixel and statistics subsections are now more carefully |
---|
453 | examined, with errors reported if there are no valid pixels in the |
---|
454 | pixel subsection, or if the statistics subsection does not intersect |
---|
455 | with the pixel subsection. |
---|
456 | |
---|
457 | Other: |
---|
458 | |
---|
459 | * The configure script has been slightly improved so that the pgsbox |
---|
460 | library is only added if PGPlot has been enabled - this is |
---|
461 | particularly necessary with the slightly changed build procedure for |
---|
462 | recent wcslib versions. |
---|
463 | |
---|
464 | * In a related note, one file - momentMap.cc - had some #include |
---|
465 | statements that failed when PGPlot (and hence pgsbox in WCSLIB) was |
---|
466 | not enabled. These were actually unnecesssary and have been removed. |
---|
467 | |
---|
468 | * There is a new input parameter - flagPlotSpectra - that allows the |
---|
469 | user to turn off the production of the spectral plots. It defaults |
---|
470 | to true (meaning the plots are produced). |
---|
471 | |
---|
472 | |
---|
473 | Duchamp version 1.1.11 (2011/03/01) |
---|
474 | ------------------------------------- |
---|
475 | |
---|
476 | Improvements: |
---|
477 | |
---|
478 | * The largest changes in this release involved improving the |
---|
479 | efficiency of the code, resulting in further speed-ups. The code |
---|
480 | affected was the low-level object-handling code, with the merging |
---|
481 | stages in particular showing good speed-ups. |
---|
482 | |
---|
483 | * The default value for the beam size, when the beam parameters are |
---|
484 | not in the FITS header, has been dropped. Now the default behaviour |
---|
485 | is to simply not make any beam corrections if the beam is not |
---|
486 | specified by either the FITS file or the input parameters. Thanks to |
---|
487 | Tobias Westemeier for the suggestion. |
---|
488 | |
---|
489 | * The reporting of the beam size has changed slightly to provide a bit |
---|
490 | more information. |
---|
491 | |
---|
492 | Bug Fixes: |
---|
493 | |
---|
494 | * The statistics section (statSec) could be larger than the image |
---|
495 | subsection (if provided), yet the pixels required would never be |
---|
496 | read in from the FITS file. The behaviour has been changed so that |
---|
497 | the statSec, if requested, is trimmed to be wholly within the image |
---|
498 | section. |
---|
499 | |
---|
500 | * It was possible to request, for instance, spectral smoothing on a 2D |
---|
501 | image. This resulted in a blank smoothed array, which led to |
---|
502 | problems in the calculations of statistics (and hence the |
---|
503 | threshold), as well as subsequent calculations. This has been fixed, |
---|
504 | so that in cases like this the smoothing is turned off (and a |
---|
505 | warning message is displayed). |
---|
506 | |
---|
507 | * There was a problem with using a MW range and subsectioning when |
---|
508 | growing objects, as a result of changes in 1.1.10. This has been |
---|
509 | fixed. |
---|
510 | |
---|
511 | * Inversion was failing when no array was allocated (not a problem for |
---|
512 | standard Duchamp, but affected its use as a library by other code). |
---|
513 | |
---|
514 | * Writing text spectra would only occur when PGPlot output was enabled |
---|
515 | in the build (it was combined with the plotting of the spectral |
---|
516 | graphs). This has been changed so that this output can be requested |
---|
517 | without requiring PGPlot. |
---|
518 | |
---|
519 | * The equation for the beam size in Sec.4.2 of the User Guide was |
---|
520 | incorrect and has been fixed. |
---|
521 | |
---|
522 | * Minor change to the configure script to improve the detection of the |
---|
523 | PGPlot library. |
---|
524 | |
---|
525 | Code development notes: |
---|
526 | |
---|
527 | * The main work has been additions and improvements to the interfaces |
---|
528 | for the Param and the Section classes, as well as new classes to |
---|
529 | manage the beam information. The efficiency changes mentioned above |
---|
530 | also affected the internals of some of the PixelInfo classes, and |
---|
531 | provided additional functions for judging the closeness of |
---|
532 | neighbouring objects. |
---|
533 | |
---|
534 | Duchamp version 1.1.10 (2010/11/15) |
---|
535 | ------------------------------------- |
---|
536 | |
---|
537 | Improvements: |
---|
538 | |
---|
539 | * The major change for this release is a greatly improved growing |
---|
540 | algorithm. The previous incarnation was very inefficient (testing |
---|
541 | all pixels in an object, most of which only have neighbours that are |
---|
542 | already in the object), and could take an unnecessarily large amount |
---|
543 | of time in the case of extended sources. The improved algorithm |
---|
544 | makes use of voxel flags to quickly identify those voxels not yet in |
---|
545 | an object and only examines them. For a certain test case |
---|
546 | (observations of extended Galactic HI), the time required reduced |
---|
547 | from about a day to about a minute! (Note that compact sources, such |
---|
548 | as HI galaxies, will not show this degree of speedup...) Many thanks |
---|
549 | to Tara Murphy for pointing out this bug and her help in |
---|
550 | diagnosing/fixing the problem. |
---|
551 | |
---|
552 | * There have been a number of further improvements to the code to |
---|
553 | speed it up, based on some profiling. The most important was the |
---|
554 | function to add pixels to a 2D object, but there were several other |
---|
555 | minor changes as well. |
---|
556 | |
---|
557 | * The minimum size of the detections can now also be specified in |
---|
558 | terms of the total number of voxels (in addition or instead of |
---|
559 | specifying the minimum spatial pixels or channels). |
---|
560 | |
---|
561 | * The way memory is allocated when reading the image has been |
---|
562 | streamlined. The amount of memory being allocated is also now |
---|
563 | reported - this is in an effort to allow users to identify problems |
---|
564 | with running out of memory. |
---|
565 | |
---|
566 | * The version number is now reported in the results and log files. |
---|
567 | |
---|
568 | * Only the filename of the input image (ie. not the leading |
---|
569 | directories) is now shown on the detection and moment-0 maps. This |
---|
570 | is to avoid problems with large directory paths being too wide for |
---|
571 | the plot. |
---|
572 | |
---|
573 | * The verification script was improved for readability (although no |
---|
574 | functional change was made). |
---|
575 | |
---|
576 | Bug fixes: |
---|
577 | |
---|
578 | * Growing was not being done when a single source was found. Now |
---|
579 | fixed. |
---|
580 | |
---|
581 | * For particular WCS setups, the RA can be interpreted as being |
---|
582 | negative, which corrupted the Source ID (ie. the IAU-format |
---|
583 | string). This has been fixed. |
---|
584 | |
---|
585 | * When the flux units were changed, problems had been seen when |
---|
586 | writing a reconstructed cube to a FITS file. Any unit change is now |
---|
587 | undone prior to writing and there should now be no problem. |
---|
588 | |
---|
589 | * Filenames with spaces in them were not handled well. New functions |
---|
590 | were included and moved to a more obvious location. These filenames |
---|
591 | now work as expected. |
---|
592 | |
---|
593 | * Spectral axes of type FREQ that had a rest frequency were not being |
---|
594 | put into velocity units. Now fixed. |
---|
595 | |
---|
596 | * Improvements to the output to screen, making sure everything is |
---|
597 | covered by the verbose flag. |
---|
598 | |
---|
599 | * Better management of the allocation of arrays. |
---|
600 | |
---|
601 | * The minSize parameter in Image was never initialised, leading to |
---|
602 | unpredictable behaviour. Now fixed. |
---|
603 | |
---|
604 | Code development notes: |
---|
605 | |
---|
606 | * New functionality includes: a function to return a vector list of |
---|
607 | voxels in an Object3D or Detection; a new generic Search function |
---|
608 | (basically a front-end to each of the individual functions); more |
---|
609 | complete copy constructors for the DataArray classes; new methods |
---|
610 | for finding the velocity widths. |
---|
611 | |
---|
612 | Duchamp version 1.1.9 (2010/03/26) |
---|
613 | ------------------------------------- |
---|
614 | |
---|
615 | Improvements: |
---|
616 | |
---|
617 | * There is now the ability to save the moment-0 map as a FITS |
---|
618 | file. This is controlled by the parameters flagOutputMomentMap and |
---|
619 | fileOutputMomentMap, which work in the same way for the mask file. |
---|
620 | |
---|
621 | * There is a new parameter flagRejectBeforeMerge that, when true, |
---|
622 | rejects sources according to the minPix/minChannels requirements |
---|
623 | before the merging and growing is done. This defaults to false (so |
---|
624 | default behaviour is as before). Thanks to Tobias Westmeier for |
---|
625 | suggesting this and the above improvement. |
---|
626 | |
---|
627 | * There is also a new parameter flagTwoStageMerging. When true (the |
---|
628 | default), newly-detected objects are merged with the first close |
---|
629 | object in the list. When false, they are simply added to the end of |
---|
630 | the list. In both cases, a second round of merging is done to pick |
---|
631 | up all close matches. |
---|
632 | |
---|
633 | * The way the beam is specified in the absence of the BMAJ/BMIN/BPA |
---|
634 | parameters has been slightly changed. The use of "beamSize" was |
---|
635 | somewhat ambiguous (it meant the number of pixels covered by the |
---|
636 | beam, not the FWHM), so new parameters have been introduced. The |
---|
637 | beam can be specified by beamArea (which replaces beamSize), or |
---|
638 | beamFWHM (the FWHM of the beam in pixels). If beamFWHM is provided, |
---|
639 | it will be used, else the beamArea will be. |
---|
640 | |
---|
641 | * There is a new parameter searchType that allows the user to choose |
---|
642 | how the searching is done. The behaviour up till now has been to |
---|
643 | search the 2D channel maps one at a time to build up the 3D |
---|
644 | detections. This is the default, and can be obtained by setting |
---|
645 | searchType=spatial. Alternatively, you can set searchType=spectral |
---|
646 | to do the searching in the 1D spectra. |
---|
647 | |
---|
648 | * Profiling of the code has led to a few changes that have sped up |
---|
649 | some calculations, particularly those involving the median. Rather |
---|
650 | than using std::sort, the code now simply uses std::nth_element, |
---|
651 | which shows a substantial speedup factor. |
---|
652 | |
---|
653 | * When using the objectList parameter, the maps will now only show the |
---|
654 | selected objects, rather than all detected objects. |
---|
655 | |
---|
656 | * There is a new verification test that looks at the "growing" |
---|
657 | functionality. |
---|
658 | |
---|
659 | Bug fixes: |
---|
660 | |
---|
661 | * When saving a mask FITS file, there was an occasional memory leak |
---|
662 | that caused Duchamp to crash (only when the cube was suitably |
---|
663 | large). This has been fixed. |
---|
664 | |
---|
665 | * The median calculation had a bug that led to an incorrect value |
---|
666 | being reported. The difference will not be great (the element of an |
---|
667 | array being reported as the median was out by one), but could be |
---|
668 | noticeable depending on the dataset. |
---|
669 | |
---|
670 | * The MW range parameters did not respond to the use of a |
---|
671 | subsection. They should refer to channel ranges in the original |
---|
672 | cube, and changes have been made to the code to reflect this. |
---|
673 | |
---|
674 | * There was an occasional error in writing out FITS keywords, due to |
---|
675 | the way comment strings were intialised. This has been improved and |
---|
676 | made more robust. |
---|
677 | |
---|
678 | * Testing with gcc-4.3.2 showed a number of compilation warnings & |
---|
679 | errors, which have been fixed. |
---|
680 | |
---|
681 | * The Gaussian smoothing kernel was too small - it is now made big |
---|
682 | enough to encompass all possible values. It was also being |
---|
683 | incorrectly scaled. This had no functional effect on the detection |
---|
684 | of objects, but for other uses was incorrect. This has been |
---|
685 | rectified. |
---|
686 | |
---|
687 | * Some outputs that weren't caught by the verbose flag now are. |
---|
688 | |
---|
689 | |
---|
690 | Code development notes: |
---|
691 | |
---|
692 | The following notes only apply if you want to use the code as a |
---|
693 | library, rather than the stand-alone package. These have generally |
---|
694 | come out of the ASKAP software development. If you just use the |
---|
695 | stand-alone package, they won't be a concern. |
---|
696 | |
---|
697 | * The API for storing detected objects has changed, making it more |
---|
698 | robust from a programming point of view. |
---|
699 | |
---|
700 | * Additional versions of some functions (and the Gaussian smoothing |
---|
701 | class) have been added - this has come out of the ASKAP development |
---|
702 | work. |
---|
703 | |
---|
704 | * A number of instances of "unsigned int" being used for loops have |
---|
705 | been changed to iterators or "size_t". There shouldn't be any |
---|
706 | functional change from this -- the code should just build more |
---|
707 | cleanly. |
---|
708 | |
---|
709 | * The Detection class now records whether the parameters have been |
---|
710 | set, and unsets that flag when objects are merged. This allows |
---|
711 | external programs to track which objects have been merged and deal |
---|
712 | with them appropriately. |
---|
713 | |
---|
714 | |
---|
715 | Duchamp version 1.1.8 (2009/07/20) |
---|
716 | ---------------------------------- |
---|
717 | |
---|
718 | Improvements: |
---|
719 | |
---|
720 | * The final source list can now be sorted on one of a range of object |
---|
721 | parameters: x-value, y-value, z-value, ra, dec, vel, w50, iflux, |
---|
722 | pflux (integrated and peak flux respectively), or snr. Simply use |
---|
723 | the new input parameter sortingParam. Its default value is vel, so |
---|
724 | the default behaviour stays the same. |
---|
725 | |
---|
726 | * The FDR method involves calculating how many pixels are |
---|
727 | correlated. This previously involved multiplying the beam area by 2, |
---|
728 | implying that 2 neighbouring channels were correlated. This factor |
---|
729 | is now a user-selectable parameter, so that if you know how many |
---|
730 | spectral channels should be correlated, you can set the parameter |
---|
731 | FDRnumCorChan and the FDR calculation will be done correctly. |
---|
732 | |
---|
733 | * Screen output now has more information on the growing threshold when |
---|
734 | the cube statistics are printed. |
---|
735 | |
---|
736 | |
---|
737 | Bug fixes: |
---|
738 | |
---|
739 | * When reading in previous detections, there was a bug causing a |
---|
740 | segmentation fault. This has been fixed and made more robust. |
---|
741 | |
---|
742 | * There was an apparent bug when writing out the reconstructed cubes |
---|
743 | to a FITS file that caused the new file to be very large. Changing |
---|
744 | the cfitsio function used has fixed this. |
---|
745 | |
---|
746 | * Using NCP projections near the equator could result in a large |
---|
747 | stream of error messages, due to coordinate transformations being |
---|
748 | done in blank regions outside of the acceptable range of world |
---|
749 | coordinates. This has been fixed. |
---|
750 | |
---|
751 | * The function that grows objects had a major bug, that stopped it |
---|
752 | working when flagAdjacent=false. This has been fixed and should work |
---|
753 | fine. |
---|
754 | |
---|
755 | * The input parameter spectraTextFile was not being read |
---|
756 | properly. This has been remedied. |
---|
757 | |
---|
758 | * The units for the integrated flux were wrong for some types of pixel |
---|
759 | flux units. This has been fixed. |
---|
760 | |
---|
761 | * A minor bug caused incorrect calculation of the 20% widths. |
---|
762 | |
---|
763 | * All calls to uint were changed to unsigned int, as some compilers |
---|
764 | had problems recognising that they were the same thing. |
---|
765 | |
---|
766 | * src/fitsHeader.cc was missing an include statement for stdlib.h, |
---|
767 | again causing some compilers to fail. |
---|
768 | |
---|
769 | |
---|
770 | Duchamp version 1.1.7 (2009/01/16) |
---|
771 | ---------------------------------- |
---|
772 | |
---|
773 | Improvements: |
---|
774 | |
---|
775 | * The mask file that can be created with flagMask can now specify |
---|
776 | which object a given detected pixel belongs to, by taking the object |
---|
777 | ID as its value. This applies if you set flagMaskWithObjectNum=true |
---|
778 | (it defaults to false, which means each detected pixel has the value |
---|
779 | 1). Non-detected pixels still have the value 0. |
---|
780 | |
---|
781 | * The names of the output FITS files are now able to be specified |
---|
782 | directly by input parameters. These will override the default names |
---|
783 | that Duchamp provides. |
---|
784 | |
---|
785 | * The Karma annotation file now specifies a font - previously it did |
---|
786 | not and so was at the mercy of whatever was specified in other |
---|
787 | annotation files being viewed at the same time. Please let me know |
---|
788 | if the font I've chosen (lucidasans-12) is not suitable. |
---|
789 | |
---|
790 | * The default flag, used in the absence of 'E', 'S' or 'N', is now '-' |
---|
791 | rather than blank. This will ensure that each column in the results |
---|
792 | file has something in it. |
---|
793 | |
---|
794 | * The configure script is somewhat more robust to non-standard |
---|
795 | systems, particularly when the X11 libraries are in unusual |
---|
796 | places. It can also deal with the slightly different way that |
---|
797 | wcslib-4.3 installs itself. |
---|
798 | |
---|
799 | * The installation part of the Makefile was altered so that the |
---|
800 | symbolic link Duchamp that points to Duchamp-1.1.7 does not have the |
---|
801 | absolute path encoded in it. This will enable the files to be moved |
---|
802 | around at a later date. |
---|
803 | |
---|
804 | Bug Fixes: |
---|
805 | |
---|
806 | * Fixed a bug that caused problems when the flux units were less than |
---|
807 | 5 characters long (e.g. 'K') |
---|
808 | |
---|
809 | * The new velocity width values were not getting proper headers in the |
---|
810 | VOTable output. This has been remedied. |
---|
811 | |
---|
812 | * Minor improvements to help non-pgplot use |
---|
813 | |
---|
814 | * Minor fix in WCS functionality. |
---|
815 | |
---|
816 | * Minor fixes to the FitsHeader class. |
---|
817 | |
---|
818 | * Improved FITS data reading, that doesn't require a nullarray |
---|
819 | array. This will reduce memory usage. |
---|
820 | |
---|
821 | * Improved subsection parsing, particularly for non-standard WCS |
---|
822 | systems. |
---|
823 | |
---|
824 | * Improving the memory allocation of the Cube class. |
---|
825 | |
---|
826 | * Fixed a bug that meant the growthThreshold values were sometimes |
---|
827 | being lost. |
---|
828 | |
---|
829 | * Improving the way the growing and merging is done, particularly in |
---|
830 | the case of non-adjacent merging. |
---|
831 | |
---|
832 | * Fixed a scaling bug in the GaussSmooth function. |
---|
833 | |
---|
834 | Other: |
---|
835 | |
---|
836 | * Numerous changes to the code that do not affect how Duchamp itself |
---|
837 | runs, but were necessary for code development with the Duchamp |
---|
838 | libraries (primarily as a part of the ASKAP code development). |
---|
839 | |
---|
840 | |
---|
841 | Duchamp version 1.1.6 (2008/09/08) |
---|
842 | ---------------------------------- |
---|
843 | |
---|
844 | New features: |
---|
845 | |
---|
846 | * The velocity widths can now be quoted in two additional ways: the |
---|
847 | full width at 50% of the peak flux (ie. the FWHM) and the full width |
---|
848 | at 20% of the peak flux (w_50 and w_20 respectively). The w_50 |
---|
849 | parameter is now quoted in the results printed to the screen |
---|
850 | (instead of w_VEL), but all three are written to the results file. |
---|
851 | |
---|
852 | * The facility now exists to bypass the searching step and go straight |
---|
853 | to the output stage using a list of previously detected objects. The |
---|
854 | list is extracted from a previously-created log file. The user can |
---|
855 | change the way the spectra are plotted (e.g. choose to plot |
---|
856 | integrated rather than peak spectra), and select a subset of the |
---|
857 | detected objects to be printed to individual postscript files. Use |
---|
858 | the input parameter usePrevious (set to true) to use this facility, |
---|
859 | and provide a list of objects using the objectList parameter |
---|
860 | (eg. 1,2,4,7-10,15) |
---|
861 | |
---|
862 | * A new parameter "growthThreshold" has been created, that goes with |
---|
863 | "threshold". If you manually specify a threshold, and you wish to |
---|
864 | grow detections, you need to manually specify a growthThreshold (in |
---|
865 | flux units). |
---|
866 | |
---|
867 | Bug fixes: |
---|
868 | |
---|
869 | * Cubes that do not have a spectral axis (such as Stokes cubes) were |
---|
870 | causing memory leaks and crashing, due to some assumptions of the |
---|
871 | existence of the spectral axis. These bugs have been fixed, so that |
---|
872 | Duchamp will run on such cubes, although the WCS transforms may not |
---|
873 | always work appropriately. Please let me know via bug reports of |
---|
874 | difficulties you encounter! |
---|
875 | |
---|
876 | * The growing function has had a few bugs fixed and is now more |
---|
877 | efficient. |
---|
878 | |
---|
879 | * The FDR method was previously overriding any manually-specified |
---|
880 | threshold. This has been fixed so that if the user specifies |
---|
881 | threshold=X, then the FDR method is not called. |
---|
882 | |
---|
883 | * Some undefined things were happening with textual output when no |
---|
884 | objects were detected. This has been fixed. |
---|
885 | |
---|
886 | * Memory leaks caused by FITS WCS irregularities have been fixed |
---|
887 | (specifically, if a FITS header had NAXIS=2 but more than two WCS |
---|
888 | axes were defined). |
---|
889 | |
---|
890 | |
---|
891 | |
---|
892 | Duchamp version 1.1.5 (2008/03/20) |
---|
893 | ---------------------------------- |
---|
894 | |
---|
895 | Bug fixes: |
---|
896 | * A major bug was discovered in the calculation of integrated |
---|
897 | fluxes. The flux was incorrectly being weighted by the spatial size |
---|
898 | of the detection, resulting in a (often considerable) overestimation |
---|
899 | of the integrated flux. This has been corrected. |
---|
900 | |
---|
901 | * The beam size calculation was also incorrect, being out by a factor |
---|
902 | of 1/ln(2). |
---|
903 | |
---|
904 | * The "make install" command previously failed to work properly if the |
---|
905 | library had not been created. This has been fixed so that the |
---|
906 | installation works even if there is no library file. |
---|
907 | |
---|
908 | New Features: |
---|
909 | * You can now provide a new flux unit string, so that the pixel fluxes |
---|
910 | will be converted to these new fluxes before any calculations are |
---|
911 | done. This is done with the newFluxUnits input parameter. The new |
---|
912 | units should be directly convertible from the old ones (given by the |
---|
913 | BUNIT header -- e.g. if BUNIT = Jy/beam, you can give newFluxUnits |
---|
914 | as mJy/beam, but not mJy). |
---|
915 | |
---|
916 | * You can now specify the precision used in the output files for flux, |
---|
917 | velocity and S/N ratio columns, via new input parameters precFlux, |
---|
918 | precVel and precSNR. These will apply to the VOTable output as well |
---|
919 | as the text table and information on the spectral plots. |
---|
920 | |
---|
921 | * The Karma annotation files previously provided just a circle at the |
---|
922 | spatial location of the detected object. You now have the option to |
---|
923 | draw a border around the detected pixels (in a similar manner to the |
---|
924 | PGPLOT moment maps). Karma does not do this perfectly, so the lines |
---|
925 | may not line up exactly with the pixel borders, so the option |
---|
926 | remains to keep using the circles. This is controlled by the |
---|
927 | parameter "annotationType", with possible options being "borders" |
---|
928 | (the default) or "circles". |
---|
929 | |
---|
930 | * There is a new function getMetadata() that loads the header and WCS |
---|
931 | data, but no pixel information. This is only of interest if you are |
---|
932 | doing code development with the Duchamp library -- it does not |
---|
933 | affect the running of Duchamp itself. |
---|
934 | |
---|
935 | Duchamp version 1.1.4 (2008/02/15) |
---|
936 | ---------------------------------- |
---|
937 | |
---|
938 | Most of the changes incorporated in this update result from ASKAP code |
---|
939 | development (that makes use of some of the Duchamp functionality), |
---|
940 | but do not actually affect the way Duchamp runs (other than, in a |
---|
941 | couple of cases, making it a bit faster). Changes that are |
---|
942 | relevant for regular users of Duchamp are as follows: |
---|
943 | |
---|
944 | * A new option has been made available (thanks to Mary Putman et al |
---|
945 | for the suggestion!) to save the spectra of all detected objects in |
---|
946 | a text file. This has the form of one column for the spectral |
---|
947 | coordinates and one column of fluxes for each object. This is |
---|
948 | enabled by the input parameters flagTextSpectra and spectraTextFile. |
---|
949 | |
---|
950 | * Slight changes to the output include he formatting of the way the |
---|
951 | detected objects are written to the log file. This meant the |
---|
952 | verification files were also updated. Also the cube statistics are |
---|
953 | not printed to the results file if a user threshold is specified (as |
---|
954 | they are not calculated!). |
---|
955 | |
---|
956 | * The determination of beam size has been made more reliable, as has |
---|
957 | the BUNIT determination. |
---|
958 | |
---|
959 | * Fixed a bug that meant source merging was not done for 2D images |
---|
960 | (not usually a problem, as the source-finding takes care of this, |
---|
961 | but came up in parallelisation development). |
---|
962 | |
---|
963 | Duchamp version 1.1.3 (2008/01/08) |
---|
964 | ---------------------------------- |
---|
965 | |
---|
966 | A minor update that just improves the build process. The location of |
---|
967 | the wcslib include files was not specified correctly, so that some |
---|
968 | machines were unable to find them. The configure script has been fixed |
---|
969 | so that the compilation should work in all cases. |
---|
970 | |
---|
971 | Some minor changes have also been made to the code in the PixelMap |
---|
972 | directory, and to param.cc, to fix warnings or errors that may appear |
---|
973 | in compilation. |
---|
974 | |
---|
975 | However, there are no changes to the actual workings of the code in |
---|
976 | this minor version, so if you have a working version of Duchamp 1.1.2 |
---|
977 | there is probably no reason to update. |
---|
978 | |
---|
979 | Duchamp version 1.1.2 (2007/12/07) |
---|
980 | ---------------------------------- |
---|
981 | |
---|
982 | Compilation: |
---|
983 | |
---|
984 | * The compilation had problems working correctly when the cfitsio and |
---|
985 | wcs libraries were not in standard locations. The configure script |
---|
986 | has been tweaked to make sure the libraries are found. |
---|
987 | |
---|
988 | * Problems with compiling in the absence of pgplot have been fixed, |
---|
989 | and it should now compile fine. |
---|
990 | |
---|
991 | * Improved the configure script for use with Mac OS X and gfortran. |
---|
992 | |
---|
993 | |
---|
994 | Bugfixing: |
---|
995 | |
---|
996 | * A major problem with the text-based output was fixed. The widths of |
---|
997 | the columns were not being calculated correctly, and were running |
---|
998 | together. This should no longer be the case. |
---|
999 | |
---|
1000 | * There was a bug in the way Duchamp interpreted spectral axes, |
---|
1001 | particularly AIPS types such as FELO-HEL, leading to incorrect |
---|
1002 | scaling of the velocity axis. This has been fixed, so that |
---|
1003 | non-standard types are now dealt with appropriately and the velocity |
---|
1004 | scaling should be accurate. |
---|
1005 | |
---|
1006 | * Another bug meant the beam correction was always applied to the |
---|
1007 | integrated spectra, regardless of whether the flux units involved |
---|
1008 | the beam (e.g. Jy/beam). This has been fixed. Note that the values |
---|
1009 | of the integrated flux presented in the results table were always |
---|
1010 | calculated correctly. |
---|
1011 | |
---|
1012 | * Other minor changes to improve the speed and efficiency, especially |
---|
1013 | of the merging code. |
---|
1014 | |
---|
1015 | New user parameters: |
---|
1016 | |
---|
1017 | * There is now the possibility to save a FITS file that contains a |
---|
1018 | mask array, indicating where the detected objects are. This is |
---|
1019 | governed by the flagOutputMask parameter. |
---|
1020 | |
---|
1021 | * The header information that is printed at the start of the results |
---|
1022 | file -- that is, the list of parameters, the statistics of the cube, |
---|
1023 | and the number of detections -- can now be written to a separate |
---|
1024 | file. Set the flagSeparateHeader parameter to true or 1 and it will |
---|
1025 | go to the file indicated by the headerFile parameter (the default is |
---|
1026 | duchamp-Results.hdr). |
---|
1027 | |
---|
1028 | * A "scaleMax" parameter has been added, so that the wavelet |
---|
1029 | reconstruction can include only a specified range of scales (for |
---|
1030 | instance, if one wants to search just for narrow features such as |
---|
1031 | absorption lines). |
---|
1032 | |
---|
1033 | * A "flagRobustStats" parameter has been added. This allows the user |
---|
1034 | to elect to use robust statistics (median and median absolute |
---|
1035 | deviation from the median) to estimate the noise parameters. If set |
---|
1036 | to false, the mean and standard deviation will be used. The default |
---|
1037 | value is true. |
---|
1038 | |
---|
1039 | New Feature: |
---|
1040 | |
---|
1041 | * When Duchamp is run on a 2D image (i.e. with no spectral output), it |
---|
1042 | now produces a postscript file that shows information each |
---|
1043 | individual source plus the image cutout of the source. This is |
---|
1044 | similar to the spectral output file, but without the spectra. The |
---|
1045 | filename is given by the spectraFile parameter. |
---|
1046 | |
---|
1047 | * Ability to save a FITS file containing a mask array, with pixel |
---|
1048 | values of 1 for pixels in a detected object, and 0 elsewhere. |
---|
1049 | |
---|
1050 | |
---|
1051 | Duchamp version 1.1.1 (2007/07/20) |
---|
1052 | ------------------------------------ |
---|
1053 | |
---|
1054 | Minor update to the release, with just a few changes: |
---|
1055 | |
---|
1056 | * The verification script was generating diffs due to differences in |
---|
1057 | the precision of some calculations when done on different machines. |
---|
1058 | One of these was the integrated flux, and this has been corrected so |
---|
1059 | that the same value should appear on all machines. |
---|
1060 | The verification script has also changed so that just the number of |
---|
1061 | sources and their positions are tested, as well as the |
---|
1062 | logfiles. This should avoid the reporting of spurious differences. |
---|
1063 | |
---|
1064 | * The grey-scale used in the image cutouts (in the spectral plots) has |
---|
1065 | been inverted. This makes it consistent with the scale used in the |
---|
1066 | maps, and hopefully easier to see when printing. |
---|
1067 | |
---|
1068 | * The configure/make system has been altered to make it easier to |
---|
1069 | install, and to specify different external libraries. In particular, |
---|
1070 | it is now not necessary to have PGPLOT installed on your system to |
---|
1071 | run Duchamp. In the absence of PGPLOT, Duchamp will run as normal, |
---|
1072 | just without any graphical output. Some source code needed to be |
---|
1073 | restructured to make this work. Please read the README file, or the |
---|
1074 | Installation appendix in the User's Guide, for complete details. |
---|
1075 | |
---|
1076 | * The time of completion is now written to the log file, so that the |
---|
1077 | user can have some idea of how long the program took. |
---|
1078 | |
---|
1079 | |
---|
1080 | Duchamp version 1.1 (2007/05/17) |
---|
1081 | ------------------------------------ |
---|
1082 | |
---|
1083 | Changes made since the last numbered release: |
---|
1084 | |
---|
1085 | EFFICIENCY OF EXECUTION: |
---|
1086 | |
---|
1087 | * Major change to the way detections are stored. They are now recorded |
---|
1088 | via run-length encoding in the x-y plane, and stored as collections |
---|
1089 | of channel-maps. This improves the memory usage and the speed of |
---|
1090 | execution (particularly for routines such as the merging |
---|
1091 | routines). The memory usage has also been improved by changing the |
---|
1092 | way the list of detected objects is stored. |
---|
1093 | |
---|
1094 | SMOOTHING: |
---|
1095 | |
---|
1096 | * Addition of ability to spatially smooth the cube before |
---|
1097 | searching. This is done with a Gaussian kernel, the size and shape |
---|
1098 | of which can be specified by the user, using the parameters kernMaj |
---|
1099 | & kernMin (which are the FWHM for the major & minor axes -- if |
---|
1100 | kernMin is not given it is assumed to match kernMaj) and kernPA (the |
---|
1101 | position angle of the major axis). There is a new parameter |
---|
1102 | smoothType that distinguishes this method from the spectral (Hanning) |
---|
1103 | smoothing already in use. |
---|
1104 | |
---|
1105 | DETECTIONS: |
---|
1106 | |
---|
1107 | * Changed the way detection is done -- now only search in 2D images, |
---|
1108 | keeping all detections (ie. no minimum-pixel requirement). This |
---|
1109 | should be the same as doing both the 1D and 2D searches with the |
---|
1110 | minimum pixel requirement in, but simpler. The rejection is done at |
---|
1111 | the end after all the merging and growing has been done. This leads |
---|
1112 | to a larger number of intermediate detections (ie. the number |
---|
1113 | before merging), which may increase the run-time slightly, although |
---|
1114 | this is offset by the changes to the efficiency (see above). |
---|
1115 | |
---|
1116 | * Increased the range over which objects can be grown to match the |
---|
1117 | threshSpatial & threshVelocity parameters, as well as rejecting new |
---|
1118 | pixels based on the BLANK and MW tests. |
---|
1119 | |
---|
1120 | INPUTS: |
---|
1121 | |
---|
1122 | * There has been a subtle change in the way BLANK pixels are dealt |
---|
1123 | with. To enable trimming of BLANK pixels, users should use the new |
---|
1124 | parameter flagTrim. The parameters flagBlankPix and blankPixVal are |
---|
1125 | no longer accessible via the parameter file -- warning messages are |
---|
1126 | given if they are provided in the parameter file. |
---|
1127 | |
---|
1128 | * Added ability to specify a statsec, so that the statistics are only |
---|
1129 | calculated from a given subsection of the cube, rather than the full |
---|
1130 | cube. |
---|
1131 | |
---|
1132 | * The subsections can also specify the ability to remove a border of n |
---|
1133 | pixels from the edges of an axis (ie. n from the start and from the |
---|
1134 | end) -- as requested in Trac ticket #5. |
---|
1135 | |
---|
1136 | * Changed the default value of the reconDim parameter from 3 to 1, and |
---|
1137 | the default value of the flagATrous parameter to false. |
---|
1138 | |
---|
1139 | OUTPUTS: |
---|
1140 | |
---|
1141 | * The output now includes three different estimates of the "centre" of |
---|
1142 | the detection: peak location, centroid (flux-weighted), or average. |
---|
1143 | |
---|
1144 | * Some slight improvements to the user feedback during execution. |
---|
1145 | |
---|
1146 | * Spectral output now includes the baseline, if one was fitted. The |
---|
1147 | format of the plot headers is slightly changed, to stop information |
---|
1148 | falling off the edge of the page. |
---|
1149 | |
---|
1150 | * Added command-line option -x to directly disable the PGPLOT X-window |
---|
1151 | map plotting. This overrides the setting of flagXOutput in the |
---|
1152 | parameter file. |
---|
1153 | |
---|
1154 | * Fallback spectral units (in case the FITS header fails to record |
---|
1155 | them) changed to SPC, but code has been fixed so that any problem |
---|
1156 | should be caught before this becomes an issue. |
---|
1157 | |
---|
1158 | BUGFIXES: |
---|
1159 | |
---|
1160 | * Fixed bug (Trac ticket #4) that meant the velocity units were not |
---|
1161 | correctly dealt with. Duchamp should be more robust for different |
---|
1162 | velocity axis setups now. |
---|
1163 | |
---|
1164 | * Fixed bug (ticket #9) with the VOTable output, where the tags were |
---|
1165 | not closing. |
---|
1166 | |
---|
1167 | * Fixed a bug that meant the integrated fluxes were not calculated |
---|
1168 | properly -- no allowance was made for the number of spatial |
---|
1169 | pixels. This should also work properly for 2D images now as well. |
---|
1170 | |
---|
1171 | * Fixed minor bug in calculating the flux threshold from the p-value |
---|
1172 | threshold when the FDR method is used. This did not affect the |
---|
1173 | actual thresholding, merely the quoted flux threshold in the results |
---|
1174 | file. Also, the MW channels are kept out of the p-value threshold |
---|
1175 | determination if they are being flagged. |
---|
1176 | |
---|
1177 | |
---|
1178 | |
---|
1179 | Duchamp version 1.0.7 (2006/11/22) |
---|
1180 | ---------------------------------- |
---|
1181 | |
---|
1182 | One new feature, and some improvements and bug-fixes. |
---|
1183 | |
---|
1184 | * The user can now elect to Hanning-smooth the spectral direction of |
---|
1185 | the cube before the searching is done. This can be done instead of the |
---|
1186 | wavelet reconstruction (although if both are selected, the |
---|
1187 | reconstruction takes precendence). This functionality is switched on |
---|
1188 | and off by the flagSmooth parameter, and the width of the Hanning |
---|
1189 | kernel is governed by the hanningWidth parameter. |
---|
1190 | |
---|
1191 | * The smoothed array can also be written to a FITS file and read in at |
---|
1192 | the start in the same manner as the reconstructed array. |
---|
1193 | |
---|
1194 | * Several improvements have been made to the memory |
---|
1195 | management. Several memory leaks were fixed, and the memory usage of |
---|
1196 | Duchamp is considerably less, which will be a help for running it on |
---|
1197 | large cubes. The speed of some routines has also been improved, by |
---|
1198 | removing unnecessarily repetitive loops. |
---|
1199 | |
---|
1200 | * The sorting routines have been changed, and should now be slightly quicker |
---|
1201 | (and a bit more robust). |
---|
1202 | |
---|
1203 | * Enabled the scale bar (plotted on the 0th moment map in the spectral |
---|
1204 | output file) to be able to deal with much finer spatial scales (as |
---|
1205 | may be expected from VLBI data). |
---|
1206 | |
---|
1207 | * Other minor changes to code and presentation. |
---|
1208 | |
---|
1209 | |
---|
1210 | Duchamp version 1.0.6 (2006/11/01) |
---|
1211 | ---------------------------------- |
---|
1212 | |
---|
1213 | Some incremental improvements and changes to the code: |
---|
1214 | |
---|
1215 | * The statistics used for the detection are now calculated once for |
---|
1216 | the entire cube, rather than being calculated separately for each |
---|
1217 | channel map and spectrum. This results in a uniform threshold value |
---|
1218 | across the entire cube. The way the statistics calculations are |
---|
1219 | implemented in the code was changed, using a new statistics class. |
---|
1220 | |
---|
1221 | * This means we can report a peak signal-to-noise value for each |
---|
1222 | detection. This is given in the text-based output, as well as in the |
---|
1223 | summary information above the spectral output for each source. |
---|
1224 | |
---|
1225 | * The user can now elect not to have the moment map displayed in a |
---|
1226 | PGPlot window, through the use of the new flagXOutput |
---|
1227 | parameter. This should help in cases of running Duchamp on a dumb |
---|
1228 | terminal (with no X-display capabilities), or as part of a batch |
---|
1229 | script where the continual popping-up of the PGPlot window might |
---|
1230 | become annoying. |
---|
1231 | |
---|
1232 | * There are different Flags reported for the case of an object lying on |
---|
1233 | the spatial edge and spectral edge of the cube (E and S respectively). |
---|
1234 | |
---|
1235 | * A few minor bug fixes: |
---|
1236 | - When providing a subsection, any axes with the range given by * |
---|
1237 | had the pixel values offset incorrectly. This has been fixed. |
---|
1238 | - Negative searches now correctly invert the reconstructed array |
---|
1239 | when it is read in from a file. |
---|
1240 | |
---|
1241 | * Other minor changes to the code, to improve memory handling and |
---|
1242 | remove redundant arrays and variables. |
---|
1243 | |
---|
1244 | |
---|
1245 | Duchamp version 1.0.5 (2006/09/06) |
---|
1246 | ---------------------------------- |
---|
1247 | |
---|
1248 | Since version 1.0, there have been a range of mostly minor |
---|
1249 | changes. These are generally bug-fixes, thanks to feedback from a |
---|
1250 | number of users, as well as a few extra features. |
---|
1251 | |
---|
1252 | Here is a summary of the changes since version 1.0: |
---|
1253 | |
---|
1254 | USER INTERFACE: |
---|
1255 | |
---|
1256 | * The user is now able to enter the value of flag parameters in the |
---|
1257 | parameter file as strings, as well as integers (so one can enter |
---|
1258 | true or 1 to give the same result). |
---|
1259 | |
---|
1260 | GRAPHICAL OUTPUT: |
---|
1261 | |
---|
1262 | * Added a new function that draws the edge of the blank region on the |
---|
1263 | maps and image cutouts. There is a new flag parameter -- |
---|
1264 | drawBlankEdges -- that controls the use of this function (the |
---|
1265 | default is true). |
---|
1266 | Also, the edge of the field is plotted on the image cutouts as a |
---|
1267 | yellow line. This distinguishes it from the purple blank pixel |
---|
1268 | dividing line. |
---|
1269 | |
---|
1270 | * The size of the tick mark in the image cutout (indicating the |
---|
1271 | spatial scale of the image) is now adaptable, so that it shows the |
---|
1272 | scale that best fits the image (ranging from 1 arcsec to 15 |
---|
1273 | degrees). (This is a change from the previous standard length of 15 |
---|
1274 | arcmin.) |
---|
1275 | |
---|
1276 | * The zoomed spectrum had its flux scale determined by all points in |
---|
1277 | the zoom box,including those flagged by the Milky Way (MW) |
---|
1278 | range. This has been fixed so that any MW channels will not affect |
---|
1279 | the flux scaling, potentially improving the appearance of objects |
---|
1280 | detected close in velocity to strong MW emission or absorption. |
---|
1281 | |
---|
1282 | * Improved Karma Annotation file output, so that files without good |
---|
1283 | WCS are dealt with appropriately. |
---|
1284 | |
---|
1285 | TEXT-BASED OUTPUT: |
---|
1286 | |
---|
1287 | * The position resolution in the IAU-format names has been increased, |
---|
1288 | so that distinct objects close to one another are more readily |
---|
1289 | distinguished. Overlapping sources at different velocities may |
---|
1290 | still have the same name, but this is a more general problem. |
---|
1291 | |
---|
1292 | * The presentation of the results in tabular format has been improved, |
---|
1293 | so that the precision of faint sources is sufficient to show their |
---|
1294 | flux, for instance, is not 0. This also applies to the information |
---|
1295 | presented above the spectral plots. |
---|
1296 | |
---|
1297 | * Improved error and warning reporting, so that the formatting and |
---|
1298 | presentation of information is clearer. |
---|
1299 | |
---|
1300 | FITS I/O RELATED: |
---|
1301 | |
---|
1302 | * The previous version was did not deal well with 4 (or greater) |
---|
1303 | dimensions in the FITS file -- the memory allocation was badly done, |
---|
1304 | and the axes needed to be in a particular order. This has been |
---|
1305 | fixed, so that the FITS I/O routines can now handle any sized FITS |
---|
1306 | file, with axes in any order (as long as there are two spatial and one |
---|
1307 | spectral axes). |
---|
1308 | |
---|
1309 | * When the FITS file does not have a rest frequency defined, there is |
---|
1310 | no way to transform the frequency axis into velocity. In this case, |
---|
1311 | the spectral axis is left as frequency, and put into units of MHz. |
---|
1312 | |
---|
1313 | * If the beam size is not indicated by the FITS header (through the |
---|
1314 | BMAJ and BMIN keywords), the user can manually specify the size |
---|
1315 | through the parameter file (using parameter beamSize). |
---|
1316 | |
---|
1317 | * The FITS I/O routines make use of the command fits_file_exists. This |
---|
1318 | was introduced into the CFITSIO package from version 2.5, so Duchamp |
---|
1319 | will only work with this version or higher. Version 3+ is preferred. |
---|
1320 | |
---|
1321 | OTHER BUG-FIXING: |
---|
1322 | |
---|
1323 | * Several bugs related to the blank-pixels were dealt with. |
---|
1324 | The correct values were not propagating through correctly, meaning |
---|
1325 | that the default values were being used. Also, there was the |
---|
1326 | potential for blank pixels to be detected when the FDR method was |
---|
1327 | used. Finally, the image cutout displays occasionally did not |
---|
1328 | obscure the blank pixels. All these problems should be fixed now. |
---|
1329 | |
---|
1330 | * The FDR setup routine was occasionally failing with a segmentation |
---|
1331 | fault, due to it accessing unallocated memory. This has been fixed. |
---|
1332 | |
---|
1333 | * Fixed bugs that affected the way 2D images (rather than 3D cubes) |
---|
1334 | were dealt with. The minChannels test failed when minChannels=0. |
---|
1335 | |
---|
1336 | * If there is no beam information in the FITS header, the user can now |
---|
1337 | enter a beam size as a parameter. |
---|
1338 | |
---|
1339 | * Other minor fixes to the code, to improve its structure and |
---|
1340 | robustness. |
---|
1341 | |
---|
1342 | |
---|
1343 | Duchamp version 1.0 (2006/06/30) |
---|
1344 | -------------------------------- |
---|
1345 | |
---|
1346 | The first full release version. |
---|
1347 | |
---|
1348 | Several major changes compared to previous version: |
---|
1349 | |
---|
1350 | * Incorporated a FitsHeader class, to keep track of header information |
---|
1351 | (including WCS) from the FITS file. All units-related calculations |
---|
1352 | are done with this class, such as conversion to velocity, and |
---|
1353 | pixel-to-WCS conversion. The main advantage of this is improved |
---|
1354 | speed of execution, more adaptable to different units, and improved |
---|
1355 | code (easier for me!). |
---|
1356 | |
---|
1357 | * Able to select the dimensionality of the reconstruction, rather than |
---|
1358 | just doing it in 3-D. |
---|
1359 | |
---|
1360 | * Robust text output, so that the columns are sized correctly given |
---|
1361 | the data being written, and the relevant units (derived from the |
---|
1362 | FITS header) are included in the output table. |
---|
1363 | |
---|
1364 | * Improved the way the reconstructed image is saved and read back in, |
---|
1365 | with a better naming scheme and a better set of new FITS header |
---|
1366 | keywords. The MW headers are now not written. |
---|
1367 | |
---|
1368 | Other, more minor changes include: |
---|
1369 | |
---|
1370 | * MW channels now not set to 0, but just ignored for searching |
---|
1371 | purposes (and for the scaling of the spectral output -- they are |
---|
1372 | plotted though and highlighted with a hashed box). |
---|
1373 | |
---|
1374 | * The scale bar in the moment map in the spectral output file has its |
---|
1375 | length indicated. |
---|
1376 | |
---|
1377 | * Improved VOTable output, with more information about the way Duchamp |
---|
1378 | was run. |
---|
1379 | |
---|
1380 | * Made sure all statistics calculations in the reconstruction are |
---|
1381 | based on robust (ie. median) methods. |
---|
1382 | |
---|
1383 | Also moved to an autoconf based configure/make compilation, for ease of use. |
---|
1384 | |
---|
1385 | |
---|
1386 | Duchamp version 0.9.2 (2006/06/27) |
---|
1387 | -------------------------------- |
---|
1388 | |
---|
1389 | * Added the use of warning flags for detections: for either edge |
---|
1390 | location or negative enclosed flux. |
---|
1391 | |
---|
1392 | * A new command line option added so that the user can specify a FITS |
---|
1393 | file to be searched with the default parameters, rather than giving |
---|
1394 | a full parameter file. |
---|
1395 | |
---|
1396 | * The scale bar on the moment cutout now works in general case, using |
---|
1397 | an iterative process to get to the correct length. |
---|
1398 | |
---|
1399 | * RA axes are now done correctly on the full-scale maps. |
---|
1400 | |
---|
1401 | * Various coding fixes, improving the readability of the code, |
---|
1402 | particularly regarding the declaration of temporary variables. |
---|
1403 | |
---|
1404 | * Changed the way the time of execution is called : now using time.h |
---|
1405 | functions, rather than a system call. |
---|
1406 | |
---|
1407 | * Changed output in Detection/outputDetection.cc for the pixel |
---|
1408 | information of detected objects (the info printed above the spectral |
---|
1409 | plots) -- now print out number of voxels in object, and have |
---|
1410 | improved the formatting a little. |
---|
1411 | |
---|
1412 | * Improved the functionality of saving the FITS file containing the |
---|
1413 | reconstructed array. Added ability to save and read back in the MW |
---|
1414 | parameters when saving a reconstructed cube (if a reconstructed cube |
---|
1415 | has different MW range to that being considered, it would cause |
---|
1416 | problems). ReadRecon now just returns a FAILURE and will reconstruct |
---|
1417 | the cube anew. |
---|
1418 | |
---|
1419 | * Changed default value of flagGrowth to 0. |
---|
1420 | |
---|
1421 | |
---|
1422 | |
---|
1423 | Duchamp version 0.9 (2006/05/05) |
---|
1424 | -------------------------------- |
---|
1425 | |
---|
1426 | First public release version. |
---|