J/A+A/679/A18 DIKL method for high contrast imaging (Ren+, 2023)
Karhunen-Loeve data imputation in high contrast imaging.
Ren B.B.
<Astron. Astrophys. 679, A18 (2023)>
=2023A&A...679A..18R 2023A&A...679A..18R (SIMBAD/NED BibCode)
ADC_Keywords: Stars, early-type ; Infrared
Keywords: circumstellar matter - quasars: general -
techniques: high angular resolution - techniques: image processing -
methods: statistical
Abstract:
Detection and characterization of extended structures is a crucial
goal in high contrast imaging. However, these structures face
challenges in data reduction, leading to over-subtraction from
speckles and self-subtraction with most existing methods. Iterative
post-processing methods offer promising results, but their integration
into existing pipelines is hindered by selective algorithms, high
computational cost, and algorithmic regularization. To address this
for reference differential imaging (RDI), here we propose the data
imputation concept to Karhunen-Loeve transform (DIKL) by modifying
two steps in the standard Karhunen-Loeve image projection (KLIP)
method. Specifically, we partition an image to two matrices: an anchor
matrix which focuses only on the speckles to obtain the DIKL
coefficients, and a boat matrix which focus on the regions of
astrophysical interest for speckle removal using DIKL components. As a
non-iterative approach, DIKL achieves high-quality results with
significantly reduced computational costs (∼3 orders of magnitude
less than iterative methods). Being a derivative method of KLIP, DIKL
is seamlessly integrable into high contrast imaging pipelines for RDI
observations.
Description:
FITS file images of circumstellar disk in Ks-band total intensity with
VLT/SPHERE/IRDIS for HD 169142 and PDS 201.
FITS file images of circumstellar disk in H-band total intensity with
VLT/SPHERE/IRDIS for HD 129590.
The stars are located at the mathematical centers of the arrays.
The units are in detector counts.
The SPHERE data were reduced with three different data reduction
methods for reference differential imaging: Karhunen-Loeve Image
Projection (KLIP), Data Imputation using Karhunen-Loeve (DIKL), and
Data Imputation using sequential Nonnegative Matrix Factorization
(DI-sNMF).
Objects:
----------------------------------------------------
RA (2000) DE Designation(s)
----------------------------------------------------
18 24 29.77 -29 46 49.3 HD 169142 = PDS 514
05 44 18.79 +00 08 40.4 PDS 201 = HIP 27059
14 44 30.96 -39 59 20.6 HD 129590 = HIP 72070
----------------------------------------------------
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
list.dat 174 13 List of fits files
fits/* . 13 Individual fits files
--------------------------------------------------------------------------------
Byte-by-byte Description of file: list.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 A9 --- Star Star name
11- 19 F9.5 deg RAdeg Right Ascension of center (J2000)
20- 28 F9.5 deg DEdeg Declination of center (J2000)
30- 36 F7.5 arcsec/pix scale ? Scale of the image
38- 40 I3 --- Nx Number of pixels along X-axis
42- 44 I3 --- Ny Number of pixels along Y-axis
46- 55 A10 "datime" Obs.date Observation date
57- 59 I3 Kibyte size Size of FITS file
61- 85 A25 --- FileName Name of FITS file, in subdirectory fits
87-174 A88 --- Title Title of the FITS file
--------------------------------------------------------------------------------
Acknowledgements:
Bin Ren, bin.ren(at)oca.eu
(End) Patricia Vannier [CDS] 30-Aug-2023