Radio Galaxy Zoo Talk

About Contour Overlay Images

  • JeanTate by JeanTate

    This post: Brief outline

    Second post: Brief description of what's in the contour overlay images

    Third post: How I produced the contour overlay images

    Fourth post: What this thread's about

    This OP is brief, because once posted, I cannot subsequently edit it (apparently)

    Posted

  • JeanTate by JeanTate

    Brief description of what's in the contour overlay images

    The background image is:

    The details of how "Luptonized" images were created are in the next post.

    The color contours are derived from FITS files produces using SkyView with Python code described in the next post.

    Unless otherwise noted, the color of the contours indicates the source:

    • red: FIRST
    • cyan: NVSS
    • lime: WISE, 3.4μ band
    • yellow: VLSSr

    RA and Dec coordinates are J2000.0.

    z_sp (z_ph) is the spectroscopic (photometric) redshift of the object in the center of the image, per SDSS (or other source, as noted). If the image is not centered on an object, or if there is no reliable redshift of a centered object, the text "z_ph N/A" is used.

    If there is a physical scale bar, in cyan, it represents (to the nearest kpc) projected distance at a redshift of z_sp/z_ph (in lime), assuming a ΛCDM cosmological model with parameter values of ΩM 0.27, ΩΛ 0.73, and H0 71 kms-1Mpc-1.

    Posted

  • JeanTate by JeanTate

    How I produced the contour overlay images

    Contours produced from FITS files (WIP; needs more work)

    FITS header: RA, Dec, size (pixels), scale, orientation. Used to produce text (scale bar, orientation vectors) overlaid on image.

    Redshift - spectroscopic, photometric, or "N/A" - an independent input.

    'Zero', a.k.a. 'sky', defined as the median flux.

    Noise is modeled using the MAD (median absolute deviation) statistic, with the 'noise floor' being an input (number of sigmas).

    Smoothing is Gaussian, using an input smoothing parameter (ndimage.gaussian_filter the Python scipy submodule)

    Contour scale factor is also input; common values are √2 and √3.

    More details - including how I came to develop this Python code - are in this RGZ Talk thread.

    "Luptonized" images

    To "Luptonize" is to produce an RGB JPEG colorized using the method described in Lupton+ 2004, "Preparing Red-Green-Blue Images from CCD Data" (arXiv preprint here).

    The three WISE bands I use are:

    • 3.4μ mapped to B,
    • 4.6μ mapped to G, and
    • 12μ mapped to R

    And the three DSS2 are:

    • B mapped to B,
    • R mapped to G, and
    • IR mapped to R.

    I use one of two methods for defining I ("intensity", some sort of measure of total flux, combining all three bands):

    • (by far the most common method) take the DN (WISE)/[?] (DSS2) values in the FITS files downloaded from SkyView
    • (a method I use rarely) normalize so that the max value in each band is the same (after removing the sky, and whatever 'noise floor' above the sky).

    The Python code I wrote, to produce the Luptonized images, is similar to that described above; in particular, the 'sky' is the median, and the 'noise floor' sigma (calculated using the MAD statistic) multiplied by a small number (typically between 1.0 and 3.5).

    For f(I) I use asinh, log, or linear.

    Here are two Luptonized images of the above field, produced using these two methods (same f(I) - asinh - same m and M, same β, etc), respectively:

    enter image description here

    enter image description here

    Posted

  • JeanTate by JeanTate

    What this thread's about ...

    I have posted many 'contour overlay images' here in RGZ Talk, for example (source):

    enter image description here

    Most, if not all, posts which contain such images also contain a 'boilerplate' slab of text, at the bottom of the post; these boilerplates are a few words about the piece of sky in the image, the parts of the electromagnetic spectrum they portray, the identity of the cental object, its redshift, the source of the data used to produce the image, and so on.

    Rather than an increasingly large chunk of text, and a growing one too (as I add features, the generic boilerplate grows too), I plan to have a really simple footnote. To the effect of "click here for details of image, including an explanation of what's in it and how it was created"^. This thread - specifically the first post in it - will be is that "here". 😃

    As my contour overlay images become more diverse (shall we say), I'll edit the posts in this thread.

    Oh, and the thread FIRST and/or NVSS contours on SDSS images contains lists of the images I've posted (and some posted by others too), with links. Despite its title, included are overlays with other than FIRST and/or NVSS contours; and on images other than SDSS ... 😉

    ^ Here it is, the New one line boilerplate (to be put at the end of all posts containing contour overlay images):

    The image(s) in this post was (were) created from sources, and using methods, described in this RGZ Talk post [link to the OP of the 'About Contour Overlay Images' thread; i.e. this thread]. The object at the center of the image is [link to SDSS Explore webpage (not sure what to do about objects which are not SDSS ones). This short sentence is omitted if there is no object at the center.]

    Posted