- Timestamp:
- May 18, 2017, 6:04:18 PM (9 years ago)
- File:
-
- 1 edited
-
trunk/doc/release.2015/ps1.detrend/detrend.tex (modified) (16 diffs)
Legend:
- Unmodified
- Added
- Removed
-
trunk/doc/release.2015/ps1.detrend/detrend.tex
r39902 r40052 11 11 12 12 \RequirePackage{color} 13 \RequirePackage{code} 13 14 \input{astro.sty} 14 %\usepackage{subcaption}15 15 %\usepackage{natbib} 16 17 \usepackage[T1]{fontenc}% (2) specify encoding 16 18 17 19 % online version may use color, but print version needs b/w … … 198 200 the metadata of exposure parameters. For the PV3 processing, large 199 201 contiguous regions were defined, and the images for all exposures 200 within that region launched for the \ ippstage{chip} stage processing.202 within that region launched for the \IPPstage{chip} stage processing. 201 203 This stage performs the image detrending (described below in section 202 204 \ref{sec:detrending}), as well as the single epoch photometry 203 205 \citep{magnier2017b}, in parallel on the individual OTA device data. 204 Following the \ ippstage{chip} stage is the \ippstage{camera} stage, in206 Following the \IPPstage{chip} stage is the \IPPstage{camera} stage, in 205 207 which the astrometry and photometry for the entire exposure is 206 208 calibrated by matching the detections against the reference catalog. 207 209 This stage also performs masking updates based on the now-known 208 210 positions and brightnesses of stars that create dynamic features (see 209 Section \ref{sec:dynamic_masks} below). The \ ippstage{warp} stage is211 Section \ref{sec:dynamic_masks} below). The \IPPstage{warp} stage is 210 212 the next to operate on the data, transforming the detector oriented 211 \ ippstage{chip} stage images onto common sky oriented images that have213 \IPPstage{chip} stage images onto common sky oriented images that have 212 214 fixed sky projections (Section \ref{sec:warping}). When all 213 \ ippstage{warp} stage processing is done for the region of the sky,214 \ ippstage{stack} processing is performed (Section \ref{sec:stacking})215 \IPPstage{warp} stage processing is done for the region of the sky, 216 \IPPstage{stack} processing is performed (Section \ref{sec:stacking}) 215 217 to construct deeper, fully populated images from the set of 216 \ ippstage{warp} images that cover that region of the sky. Beyond the217 \ ippstage{stack} stage, a series of additional stages are done that218 \IPPstage{warp} images that cover that region of the sky. Beyond the 219 \IPPstage{stack} stage, a series of additional stages are done that 218 220 are more fully described in other papers. Transient features are 219 identified in the \ ippstage{diff} stage, which takes input220 \ ippstage{warp} and/or \ippstage{stack} data and performs image221 identified in the \IPPstage{diff} stage, which takes input 222 \IPPstage{warp} and/or \IPPstage{stack} data and performs image 221 223 differencing (Section \ref{sec:diffs}). Further photometry is 222 performed in the \ ippstage{staticsky} and \ippstage{skycal} stages,224 performed in the \IPPstage{staticsky} and \IPPstage{skycal} stages, 223 225 which add extended source fitting to the point source photometry of 224 objects detected in the \ ippstage{stack} images, and calibrate the225 results against the reference catalog. The \ ippstage{fullforce} stage226 takes the catalog output of the \ ippstage{skycal} stage, and uses the226 objects detected in the \IPPstage{stack} images, and calibrate the 227 results against the reference catalog. The \IPPstage{fullforce} stage 228 takes the catalog output of the \IPPstage{skycal} stage, and uses the 227 229 objects detected in that to perform forced photometry on the 228 individual \ ippstage{warp} stage images. The details of these stages230 individual \IPPstage{warp} stage images. The details of these stages 229 231 are provided in \citet{magnier2017b}. 230 232 … … 234 236 the summit to the main IPP processing cluster at the Maui Research and 235 237 Technology Center in Kihei, and registered into the processing 236 database. This triggers a new \ ippstage{chip} stage reduction for238 database. This triggers a new \IPPstage{chip} stage reduction for 237 239 science exposures, advancing processing upon completion through to the 238 \ ippstage{diff} stage. This allows the ongoing solar system moving240 \IPPstage{diff} stage. This allows the ongoing solar system moving 239 241 object search to identify candidates for follow up observations within 240 242 24 hours of the initial set of observations \citep{2015IAUGA..2251124W}. … … 244 246 details of the construction of those detrends in Section 245 247 \ref{sec:detrend construction}. An analysis of the algorithms used to 246 complete the \ ippstage{warp} (section \ref{sec:warping}),247 \ ippstage{stack} (section \ref{sec:stacking}), and \ippstage{diff}248 complete the \IPPstage{warp} (section \ref{sec:warping}), 249 \IPPstage{stack} (section \ref{sec:stacking}), and \IPPstage{diff} 248 250 (section \ref{sec:diffs}) stage transformations of the image data to 249 251 from the detector frame to a common sky frame, and the co-adding of … … 297 299 case with GPC1, and this requires an additional set of detrending 298 300 steps to remove these non-linear responses. The first of these is the 299 \ ippprog{burntool} correction, which removes the persistence trails301 \IPPprog{burntool} correction, which removes the persistence trails 300 302 caused by the incomplete transfer of charge along the readout columns. 301 303 This bright-end nonlinearity is generally only evident for the … … 321 323 322 324 For the PV3 processing, all detrending is done by the 323 \ ippprog{ppImage} program. This program applies the detrends to the325 \IPPprog{ppImage} program. This program applies the detrends to the 324 326 individual cells, and then an OTA level mosaic is constructed for the 325 327 science image, the mask image, and the variance map image. The single … … 354 356 355 357 Both of these types of persistence trails are measured and optionally 356 repaired via the \ ippprog{burntool} program. This program does an358 repaired via the \IPPprog{burntool} program. This program does an 357 359 initial scan of the images, and identifies objects with pixel values 358 360 brighter than a conservative threshold of 30000 DN. The trail from … … 716 718 models for each individual dark model. The additional pixel-to-pixel 717 719 variance from this noisemap is added to the Poissonian variance to 718 form the science variance image generated by the \ ippstage{chip}720 form the science variance image generated by the \IPPstage{chip} 719 721 processing. 720 722 … … 1064 1066 1065 1067 The remaining dynamic masks are not generated until the IPP 1066 \ ippstage{camera} stage, at which point all object photometry is1068 \IPPstage{camera} stage, at which point all object photometry is 1067 1069 complete, and an astrometric solution is known for the exposure. This 1068 1070 added information provides the positions of bright sources based on … … 1261 1263 \label{sec:masking_fraction} 1262 1264 1263 For the full field of view that falls on the sixty OTAs, 14.7\% of all 1264 pixels are masked. The large fraction of this masking is due to 1265 regions that fall within the vignetted region. Defining the diameter 1266 of the unvignetted region to have be 3 degrees, and excluding pixels 1267 that fall beyond this point reduces the static masking fraction to 1268 9.7\%. 1269 1270 Unfortunately, due to the design of the OTAs and readout cells, a 1271 non-negligible fraction of the field of view falls onto an area that 1272 does not have a detector pixel. For a given OTA mosaicked to a 1273 $4846\times{}4868$ pixel image, the 64 $590\times{}598$ pixel readout 1274 cells cover 95.7\% of the OTA area, providing an additional 4.3\% 1275 masking in the unvignetted field of view due to the absence of a 1276 detector pixel. 1277 1278 For the inter-chip gap area loss, we use two field of view 1279 calculations to estimate the masking fraction. The reference field of 1280 view of GPC1 is 3 degrees, which at the nominal plate scale of 0.258 1281 arcseconds per pixel, translates to a 20930 FPA pixel radius. Summing 1282 mask fractions from these three contributions within the unvignetted 1283 field of view results in an average of $\sim 20\%$ masking fraction 1284 across the field of view. Dynamic masking adds an additional $2-3\%$ 1285 on average, with advisory burntool masking contributing the largest 1286 single component. Table \ref{tab:mask fraction} contains estimates of 1287 the mask fraction in the GPC1 detector footprint by the sources of the 1288 masking for the 3 degree field of view, as well as for a larger 3.25 1289 degree field of view that allows addition unvignetted regions in the 1290 corners to contribute. 1265 Although there are a large number of masked pixels within the sixty 1266 OTAs of GPC1, the camera was designed to move chips with problematic 1267 areas (most notably CTE issues) to the edges of the detector. Because 1268 of this, the main analysis of the mask fraction is based not on the 1269 total footprint of the detector, but upon a circular reference field 1270 of view with a radius of 1.5 degrees. This field of view corresponds 1271 approximately to half the width and height of the detector. This 1272 field of view underestimates the unvignetted region of GPC1. A second 1273 ``maximum'' field of view is also used to estimate the mask fraction 1274 within a larger 1.628 degree radius. This larger radius includes far 1275 larger missing fractions due to the circular regions outside region 1276 populated with OTAs, but does include the contribution from 1277 well-illuminated pixels that are ignored by the reference radius. 1278 1279 The results of simulating simulating the footprint of the detector as 1280 a grid of uniformly sized pixels of $0\farcs{}258$ size are provided 1281 in Table \ref{tab:mask fraction}. Both fields of view contain 1282 circular segments outside of the footprint of the detector, which 1283 increase the area estimate that is unpopulated. This category also 1284 accounts for the inter-OTA and inter-cell gaps. The regions with poor 1285 CTE also account for a significant fraction of the masked pixels. The 1286 remaining mask category accounts for known bad columns, cells that do 1287 not calibrate well, and vignetting. There are also a small fraction 1288 that have static advisory masks marked on all images. These masks 1289 mark regions where bright columns on one cell periodically create 1290 cross talk ghosts on other cells. 1291 1292 During the \IPPstage{camera} processing, a separate estimate of the 1293 mask fraction for a given exposure is calculated by counting the 1294 fraction of pixels with static, dynamic, and advisory mask bits set 1295 within the two field of view radii. The static mask fraction is then 1296 augmented by an estimate of the unpopulated inter-chip gaps (as the 1297 input masks already account for the inter-cell gaps). This estimate 1298 does not include the circular segments outside of the detector 1299 footprint. This is minor for the reference field of view (1\% 1300 difference), but does underestimate the static mask fraction for the 1301 maximum radius by 7.3\%. This analysis does provide a the observed 1302 dynamic and advisory mask fractions, which are 0.03\% and 3\% 1303 respectively. The significant advisory value is a result of applying 1304 such masks to all burntool corrected pixels. 1291 1305 1292 1306 \begin{deluxetable}{lcc} … … 1296 1310 \tablehead{\colhead{Mask Source}&\colhead{3 Degree FOV}&\colhead{3.25 Degree FOV}} 1297 1311 \startdata 1298 No pixel & 4.44\% & 9.47\% \\ 1299 Detector defect & 6.37\% & 7.91\% \\ 1300 CTE issue & 2.62\% & 3.13\% \\ 1312 Good pixel & 78.9\% & 71.1\% \\ 1313 Unpopulated & 13.1\% & 19.6\% \\ 1314 CTE issue & 2.3\% & 2.6\% \\ 1315 Other issue & 5.4\% & 6.4\% \\ 1316 Static advisory & 0.3\% & 0.3\% \\ 1301 1317 \enddata 1302 1318 \label{tab:mask fraction} … … 1413 1429 can have their own features modeled as being part of the background. 1414 1430 For the specialized processing of M31, which covers an entire pointing 1415 of GPC1, the measured background was added back to the \ ippstage{chip}1431 of GPC1, the measured background was added back to the \IPPstage{chip} 1416 1432 stage images, but this special processing was not used for the large 1417 1433 scale $3\Pi$ PV3 reduction. … … 1422 1438 The various detrends for GPC1 are constructed in similar ways. A 1423 1439 series of appropriate exposures is selected from the database, and 1424 processed with the \ ippprog{ppImage} program. This program is used1425 for the \ ippstage{chip} stage processing as well, and is designed to1440 processed with the \IPPprog{ppImage} program. This program is used 1441 for the \IPPstage{chip} stage processing as well, and is designed to 1426 1442 do multiple image processing operations. The extent of this 1427 1443 processing is dependent on the order in which the detrend to be … … 1431 1447 for the detrends we construct. 1432 1448 1433 Once the input data has been prepared, the \ ippprog{ppMerge} program1449 Once the input data has been prepared, the \IPPprog{ppMerge} program 1434 1450 is used to construct some sort of ``average'' of the inputs. This 1435 1451 step need not be a mathematical average, but is used to combine the … … 1557 1573 1558 1574 For each output skycell, all overlapping OTAs and the calibrated 1559 catalog are read into the \ ippprog{pswarp} program. Each input image1575 catalog are read into the \IPPprog{pswarp} program. Each input image 1560 1576 is examined in order, and the same transformation performed. This 1561 1577 transformation breaks the output warp image into $128\times{}128$ … … 1666 1682 The stacked image is comprised of all warp frames for a given skycell 1667 1683 in a single filter. The source catalogs and image components are 1668 loaded into the \ ippprog{ppStack} program to prepare the inputs and1684 loaded into the \IPPprog{ppStack} program to prepare the inputs and 1669 1685 stack the frames. 1670 1686
Note:
See TracChangeset
for help on using the changeset viewer.
