Embed Size (px)
Citation preview
Processamento de Imagens
Daniel Nicolato E. Pereira
I WCCA
arX
iv:1
011.
0731
v1 [
astro
-ph.
HE]
2 N
ov 2
010
Submitted to ApJPreprint typeset using LATEX style emulateapj v. 11/10/09
USING THE X-RAY MORPHOLOGY OF YOUNG SUPERNOVA REMNANTS TO CONSTRAIN EXPLOSIONTYPE, EJECTA DISTRIBUTION, AND CHEMICAL MIXING
Laura A. Lopez1, Enrico Ramirez-Ruiz1, Daniela Huppenkothen2. Carles Badenes3,4, David A. Pooley5
Submitted to ApJ
ABSTRACT
Supernova remnants (SNRs) are a complex class of sources, and their heterogeneous nature hashindered the characterization of their general observational properties. To overcome this challenge,in this paper, we use statistical tools to analyze the Chandra X-ray images of Galactic and LargeMagellanic Cloud SNRs. We apply two techniques, a power-ratio method (a multipole expansion) andwavelet-transform analysis, to measure the global and local morphological properties of the X-ray lineand thermal emission in twenty-four SNRs. We find that Type Ia SNRs have statistically morespherical and mirror symmetric thermal X-ray emission than core-collapse (CC) SNRs. The abilityto type SNRs based on thermal emission morphology alone enables, for the first time, the typingof SNRs with weak X-ray lines and those with low resolution spectra. Based on our analyses, weidentify one source (SNR G344.7!0.1) as originating from a CC explosion that was of unknown originpreviously; we also confirm the tentative Type Ia classifications of G337.2!0.7 and G272.2!3.2.Although the global morphology is indicative of the explosion type, the relative morphology of theX-ray line emission within SNRs is not: all sources in our sample have well-mixed ejecta, irrespectiveof stellar origin. In particular, we find that 90% of the bright metal-line emitting substructures arespatially coincident and have similar scales, even if the metals arise from di!erent burning processes.Moreover, the overall X-ray line morphologies within each SNR are the same, with <6% di!erences.These findings reinforce observationally that hydrodynamical instabilities can e"ciently mix ejecta inType Ia and CC SNRs. The only exception is W49B, which can be attributed to its jet-driven/bipolarexplosive origin. Based on comparative analyses across our sample, we describe several observationalconstraints that can be used to test hydrodynamical models of SNR evolution; notably, the fillingfactor of X-ray emission decreases with SNR age.Subject headings: methods: data analysis — supernova remnants — techniques: image processing —
X-rays: ISM
1. INTRODUCTION
Supernova remnants (SNRs) are a diverse class of ob-jects that play an essential role in the Universe, in-cluding driving the dynamics of the interstellar medium(ISM) and producing and distributing most of the met-als (Fukugita & Peebles 2004). The morphology and dy-namics of young SNRs depend on the distribution ofthe ambient medium and on the structure of the stel-lar ejecta. Self-similar, spherically-symmetric solutionsexist (Chevalier 1982), and they are used widely to in-terpret observational data of young SNRs. However, theejecta are subject to hydrodynamical instabilities thatpreclude a self-similar description of the expansion, andthus the use of hydrodynamical models is necessary.A major di"culty at present is bridging these hydrody-
namical models with observations of young SNRs. A fewdirect observables of individual sources (e.g., expansionrates) can be compared easily to theoretical predictions.
1 Department of Astronomy and Astrophysics, Univer-sity of California Santa Cruz, 159 Interdisciplinary SciencesBuilding, 1156 High Street, Santa Cruz, CA 95064, USA;[email protected].
2 Astronomical Institute, Anton Pannekoek, University of Am-sterdam, P.O. Box 94249, 1090 GE Amsterdam, Netherlands
3 School of Physics and Astronomy, Tel-Aviv University, Tel-Aviv 59978, Israel
4 Benoziyo Center for Astrophysics, Weizmann Institute ofScience, Rehovot 76100, Israel
5 Eureka Scientific, Inc., Austin, TX 78756
However, the complexity and heterogeneous nature ofSNRs has limited the ability to define the observed prop-erties of SNRs as a class. As a consequence, previousobservational SNR work has largely focused on interpre-tation of single objects, without systematic comparisonbetween sources (with some exceptions: e.g., Badeneset al. 2010; Long et al. 2010). Although each SNR isunique and complicated when studied in detail, it is vitalto unify the observed characteristics of SNRs to test andto improve hydrodynamical models of their dynamics.With the advent of high-resolution, space-based tele-
scopes in the last couple decades, the time is ripe toundertake this task. In particular, the Chandra X-rayObservatory has facilitated an unprecedented view ofyoung ejecta-dominated SNRs since its launch in 1999.The sub-arcsecond spatial resolution and the spatially-resolved spectroscopy capabilities of Chandra have facil-itated detailed studies of the metal-rich ejecta from SNexplosions as well as their interactions with the surround-ing as they expand (see reviews by Weisskopf & Hughes2006; Badenes 2010). Chandra has observed over one-hundred SNRs in the Milky Way galaxy (Green 2009)and many others in nearby galaxies (e.g., M33: Long etal. 2010). This wealth of data provides the necessary ba-sis to characterize the observed X-ray properties of SNRsas a class.Toward this end, in this paper, we use quantitative
methods to examine the Chandra images of all SNRs
X-ray Morphological Properties of SNRs 13
APPENDIX
A.1. WAVELET-TRANSFORMED IMAGES
Fig. 8.— Raw images of line emission (O continuum, Mg xi, Si xiii, S xv, Ar xvii, Ca xix, and Fe xxv) in Cas A and correspondingwavelet-transformed images for five di!erent scales. The white scale bar is 1’ ! 1 pc in length. The color bar is set so blue is the minimum,and red is the maximum.
A&A 526, A144 (2011)DOI: 10.1051/0004-6361/201014358c! ESO 2011
Astronomy&Astrophysics
High-contrast optical imaging of companions:the case of the brown dwarf binary HD 130948 BC
L. Labadie1 ,2, R. Rebolo1 ,6, I. Villó3, J. A. Pérez-Prieto1, A. Pérez-Garrido3, S. R. Hildebrandt4, B. Femenía1,2,A. Díaz-Sanchez3, V. J. S. Béjar1,2, A. Oscoz1, R. López1, J. Piqueras5, and L. F. Rodríguez1
1 Instituto de Astrofisica de Canarias, C/ Via Lactea s/n, La Laguna, 38200 Tenerife , Spaine-mail: [email protected]
2 Departamento de Astrofisica, Universidad de La Laguna, 38205 La Laguna, Tenerife, Islas Canarias, Spain3 Universidad Politecnica de Cartagena, Campus Muralla del Mar, Cartagena, 30202 Murcia, Spain4 Laboratoire de Physique Subatomique et de Cosmologie, 53 avenue des Martyrs, 38026 Grenoble, France5 Max-Planck-Institut für Sonnensystemforschung, Max-Planck-Str. 2, 37191 Katlenburg-Lindau, Germany6 Consejo Superior de Investigaciones Cientificas, Spain
Received 5 March 2010 / Accepted 27 September 2010
ABSTRACT
Context. High-contrast imaging at optical wavelengths is limited by the modest correction of conventional near-IR optimized AO sys-tems. We take advantage of new fast and low-readout-noise detectors to explore the potential of fast imaging coupled to post-processing techniques to detect faint companions of stars at small angular separations.Aims. We have focused on I-band direct imaging of the previously detected brown dwarf binary HD 130948 BC, attempting to spa-tially resolve the L2+L2 system considered as a benchmark for the determination of substellar objects dynamical masses.Methods. We used the lucky-imaging instrument FastCam at the 2.5-m Nordic Telescope to obtain quasi di!raction-limited images ofHD 130948 with "0.1## resolution. In order to improve the detectability of the faint binary in the vicinity of a bright (I = 5.19 ± 0.03)solar-type star, we implemented a post-processing technique based on wavelet transform filtering of the image, which allows us tostrongly enhance the presence of point-like sources in regions where the primary halo generally dominates.Results. We detect for the first time the binary brown dwarf HD 130948 BC in the optical band I with a SNR " 9 at 2.561## ± 0.007##(46.5 AU) from HD 130948 A and confirm in two independent datasets (2008 May 29 and July 25) that the object is real, as opposedto time-varying residual speckles. We do not resolve the binary, which can be explained by astrometric results posterior to our ob-servations, which predict a separation below the telescope resolution. We reach a contrast of "I = 11.30 ± 0.11 at this distance, andestimate a combined magnitude for this binary I = 16.49 ± 0.11 and a I $ J color of 3.29 ± 0.13. At 1##, we reach a detectability10.5 mag fainter than the primary after image post-processing.Conclusions. We obtain on-sky validation of a technique based on speckle imaging and wavelet-transform post-processing, whichimproves the high-contrast capabilities of speckle imaging. The I $ J color measured for the BD companion is slightly bluer, but stillconsistent with what is typically found for L2 dwarfs ("3.4–3.6).
Key words. instrumentation: high angular resolution – methods: observational – techniques: image processing – binaries: close –brown dwarfs – circumstellar matter
1. Introduction
A direct determination of dynamical masses of very low-mass(VLM) objects is essential to calibrate the mass-luminosity re-lationship. This is particularly relevant for understanding thebrown dwarfs (BDs) evolution. Dynamical masses can be deter-mined by observing close multiple BD systems (Zapatero Osorioet al. 2004; Bouy et al. 2004; Stassun et al. 2006; Dupuy et al.2009). Brown dwarfs close binaries with orbital periods <"10 yrrepresent a valuable sample for a model-independent mass deter-mination within a realistic time baseline. Observationally, thisrequires to spatially resolve the binary, which also permits usto obtain a direct measurement of the flux of each component.Because BD systems are also detected as close companionsto bright main-sequence stars, another di#culty resides in thestrong contrast needed to detect them (cf. the case HR 7672 Bin Liu et al. 2002), on top of the detectability issue becauseof their intrinsic low luminosity. So far, the sample of such
companion BD binaries is limited to a few (Burgasser et al.2005), mostly characterized in the near-IR with the help of 8–10 m class telescopes.
Optical data are necessary for a full characterization of thespectral energy distribution, which is key to the determinationof e!ective temperatures and bolometric luminosity. In the visi-ble domain, close binaries can be spatially resolved with speckleimaging (Law et al. 2006a), a technique that delivers di!raction-limited optical counterpart to AO-assisted infrared images. Thequestion of high contrast in speckle imaging has been inves-tigated in the past by Boccaletti et al. (2001) using the “darkspeckles” method as an additional stage of cleaning to improvethe detectability of faint companions. Coupled to the adaptiveoptics system ADONIS and a Lyot stellar coronograph, these au-thors obtained K-band contrasts of 1.5$4.5% 10$3 ("mK " 6$7)at 0.5–0.9##.
In this paper, we have focused on the brown dwarf binaryHD 130948 BC, originally reported by Potter et al. (2002). As
Article published by EDP Sciences A144, page 1 of 8
L. Labadie et al.: High contrast optical imaging of the brown dwarf binary HD 130948 BC
Fig. 1. Upper-left: original H-band detection of HD 130948 BC in Potter et al. (2002). Upper-right: direct lucky-imaging image resulting from thebest 30% frames over a 105 serie. The white arrow indicates the anticipated position of the companion. The inset shows the core of the PSF on thefull intensity scale. Bottom-left: post-processed image from July 2008 revealing the BD binary companion, unresolved with the NOT at 0.8 µm.Bottom-right: post-processed image from May 2008 obtained with 5 ! 104 frames. All four images have the same size. North is up, East is left.The e!ective total integration time to obtain the July image is 900 s.
and therefore, in order to correct for the di!erence, we estimatedthe aperture correction from 5 to 10 FWHM with bright andisolated Landolt standard stars. We obtained an aperture correc-tion of 0.033 ± 0.005 mag, which was included in the photom-etry of our target. Weather conditions during our observationswere photometric as assessed by observing photometric standardthrough the whole night, while the average seeing ranged from1.3 to 2"". In order to transform our instrumental magnitudesinto apparent magnitudes, we observed four di!erent Landoltstandard star fields (each of them containing 3–6 standard stars)and repeated them during the night. We obtained 14 di!erentimages at 7 di!erent pointings covering a range of airmassesfrom 1.1 to 2.1. We perform a linear fit to our data to obtain thezero points and the extinction coe"cient, following the equationi # I = a0 + k ! airmass, where i and I are, respectively, the in-strumental magnitude and the apparent magnitude of the Landoltstars, a0 is the zero point, k is the extinction coe"cient. We ob-tained a0 = 2.68± 0.019 (zmag = 25) and k = 0.17± 0.013. Theerror bars in the calibration were obtained from the estimatederrors of the coe"cients in the linear fit. Eventually, our correc-tion from instrumental to apparent magnitudes for HD 130948was finally i # I = 2.860 ± 0.032 (#0.033 ± 0.005 aperture cor-rection ) = 2.827 ± 0.032. The final error bar includes both theerror in the calibration and in the instrumental magnitude.
3. Results
3.1. Detection of the BD companion
The panels of Fig. 1 show the imaging results from our obser-vations. In the upper-left corner we display the original detec-tion by Potter et al. (2002), in which the brown dwarf binary isresolved in the H-band with the 8-m Gemini-North telescope.The upper-right corner shows the direct shift-and-add image ofHD 130948 obtained with the data of July 25. In all the images,North is up, East is left. The FastCam images have been rotatedby 90$ with respect to the original position of the detector on sky.The average full-width-at-half-maximum (FWHM) is 131 mas,with a slight elongation toward the east (148 mas) against thenorth (114 mas). We attribute this e!ect to atmospheric disper-sion because no ADC (atmospheric dispersion compensator) isavailable in the current FastCam+NOT configuration. The whitearrow indicates the expected position of the BD companion withrespect to HD 130948 A, while the inset shows the core of thePSF on a di!erent intensity scale.
The bottom part of Fig. 1 shows the HD 130948 system ob-served in July (left) and May 2008 (right) after the image fil-tering step. The brown dwarf is detected with a signal-to-noiseratio (SNR) % 9 and an average FWHM % 110 mas, at a posi-tion very consistent with earlier image (see next section). The
A144, page 3 of 8
Coronal Mass Ejection Detection using Wavelets,Curvelets and Ridgelets: Applications for Space
Weather Monitoring
P.T. Gallaghera,∗, C.A. Young
b,∗, J.P. Byrne
a, R.T.J. McAteer
a
aAstrophysics Research Group, School of Physics, Trinity College Dublin, Dublin 2,Ireland
bADNET Systems, Inc., NASA Goddard Space Flight Center, Greenbelt, MD 20850, USA
Abstract
Coronal mass ejections (CMEs) are large-scale eruptions of plasma and
magnetic field that can produce adverse space weather at Earth and other lo-
cations in the Heliosphere. Due to the intrinsic multiscale nature of features
in coronagraph images, wavelet and multiscale image processing techniques
are well suited to enhancing the visibility of CMEs and supressing noise.
However, wavelets are better suited to identifiying point-like features, such
as noise or background stars, than to enhancing the visibility of the curved
form of a typical CME front. Higher order multiscale techniques, such as
ridgelets and curvelets, were therefore explored to characterise the morphol-
ogy (width, curvature) and kinematics (position, velocity, acceleration) of
CMEs. Curvelets in particular were found to be well suited to characterising
CME properties in a self-consistent manner. Curvelets are thus likely to be
of benefit to autonomous monitoring of CME properties for space weather
applications.
Keywords:Coronal mass ejection (CME), Multiscale methods, Space weather, Wavelets
∗Corresponding AuthorsEmail addresses: [email protected] (P.T. Gallagher),
[email protected] (C.A. Young)
Preprint submitted to Advances in Space Research December 10, 2010
arX
iv:1
012.
1901
v1 [
astro
-ph.
SR]
9 D
ec 2
010
Figure 6: Filtered raw images using an isotropic wavelet (left) and curvelets (right) byremoving coefficients most probably due to noise.
Figure 7: Contrast enhanced raw images by amplifying the isotropic wavelets coefficients(left) and the curvelet coefficients (right) at the finer scales.
15
Roteiro
• Imagens digitais astronômicas
• Operações básicas
• Representações da imagem
• Análise e visão multi-escalas
Imagens Digitais Astronômicas
Peculiaridades
• Imagem esparsa
• Objetos positivos
• Objetos sem limites nítidos
Peculiaridades
• Imagem esparsa
• Objetos positivos
• Objetos sem limites nítidos
• Compressão
• Caracterização do fundo
• Determinação de regiões de interesse
• Segmentação
Baixa Intensidade
Notação
Ruído
• Ruído Poisson
• “Experimento”: emissão de fóton
• “Sucesso”: detecção de fóton
Ruído
• Ruído Gaussiano
• Ruído Misto
Ruído• Establização da Variância
• Transformação de Anscombe (1948)
...e generalizações
Fontes pontuais e PSF
Seeing + Difração
Fontes pontuais e PSF
Operações básicas
DFT
DFT
Exemplos por Fred Weinhaus em www.imagemagick.org
Convolução
Filtros Lineares
Filtros Lineares
Deconvolução
?
Deconvolução
• Wiener
• Richardson-Lucy
Representações da Imagem
Espaços e Bases
• Diferentes formas de se descrever, matematicamente, uma imagem
• Ortonormalidade...
Espaço Direto
• Bases do tipo delta
• Coeficientes são valores dos píxeis
Espaço de Fourier
• Bases combinam senóides orientados nos dois eixos
• Coeficientes formam transformada de Fourier, e expressam importância de cada freqüência
Espaço de Wavelets• Bases consistem em
famílias de funções que representam “ondas localizadas”
• Várias possibilidades
• Coeficientes formam transformada Wavelet, e expressam importância de cada escala em cada posição na imagem
Wavelets
finito
Wavelets
À Trous
-4 -3 -2 0 +1-1 +2 +3 +4
-4 -3 -2 0 +1-1 +2 +3 +4
-4 -3 -2 0 +1-1 +2 +3 +4
-4 -3 -2 0 +1-1 +2 +3 +4
À Trous
Multi-escalas em Cascata
Análise e Visão Multi-escalas
Tratamento de Ruído
Segmentação1
2
3
4
Reconstrução
Reconstrução
