-- Public.AnishaVadher - 06 Feb 2017

Scaling Correction

  • Transforming from RA/DEC into pixel coordinates:
  • CodeCogsEqn_2.gif
  • where M^-1 [pixels/degrees] is the inverse of the transformation matrix given in the header of the fits file of the image along with the reference coordinates in pixels and degrees

  • Transformation matrix applied to 5 stars in frame:
    Transformation5stars.png
  • Multiplied the horizontal transformation by factor n:
    CodeCogsEqn_3.gif
  • Set n = 1.85 to match up stars
  • Stars matched up with accurate pixel coordinates:
    Matched5stars.png
  • Table of stars plotted on diagram
Star RA/DEC [Deg] V Pixel V Mag B Pixel B Mag
1 35.480983, 57.242918 912.56, 176.98 6.39 917.95, 173.75 6.97
2 35.722909, 57.245152 138.28, 141.41 7.45 143.55, 137.99 8.16
3 35.601206, 57.109481 503.97, 956.29 7.82 509.01, 952.82 10.09
4 35.754022, 57.121345 16.74, 871.90 11.93 21.72, 868.32 12.39
5 35.380351, 57.103321 1210.09, 1013.19 10.99 1215.22, 1009.87 11.11

  • Next step - find stars n accociated magnitudes from star finding algorithm
Purity and Efficiency of Star-Finding Algorithm

purity_efficiency_eqs.png

  • Purity plotted against efficiency for single simulated images of the Double Cluster and stacked simulations (both with flat background) for 1.5<n<10
  • Expected 494 stars in the stacked frames
  • Expected 769 stars in the single frames
  • Couldn't get to a threshold lower than n=4 for stacked frames (whole image was flooded)
  • Stacked frames have improved purity and efficiency over single frames, will choose to stack frames for real data
  • DBCL_ROC_sim_vis_comp.png
  • DBCL_ROC_sim_blue_comp.png
Stacking Non-aligned Images (Drizzle Algorithm)

  • Images of the Dragonfly cluster from observation 2 are not aligned.
  • ImageOverlapDiagram1.png
  • Pixels are not aligned, therefore cannot sum pixel values directly.
  • Can use Drizzle algorithm to stack images.
  • PixelOverlapDiagram.png
  • When adding input image, new pixel value is determined by weighted fill of overlapping pixels. For each input pixel:
 \begin{equation}  \textup{ADU}_{\textup{fill}}=\textup{ADU}_{\textup{input}} \times \frac{\textup{Area}\left ( \textup{overlap} \right )}{\textup{Area}\left ( \textup{input pixel} \right )}  \end{equation}
  • Additional note: brightest star in simulated double cluster doesn't exist?
  • Real stars match up well vertically with simulated data without it in comparison to with it (shown in 20170119 meeting)

  • Real Double Cluster without brightest star data:
    RealDoubleHR.png
Topic attachments
I Attachment Action Size Date Who Comment
GIFgif CodeCogsEqn_2.gif manage 3.0 K 07 Feb 2017 - 15:10 AnishaVadher  
GIFgif CodeCogsEqn_3.gif manage 1.8 K 07 Feb 2017 - 15:16 AnishaVadher Multiplied the horizontal transformation by factor n
PNGpng DBCL_ROC_sim_blue_comp.png manage 47.7 K 15 Feb 2017 - 15:43 LauraBevilacqua ROC curves
PNGpng DBCL_ROC_sim_vis_comp.png manage 47.0 K 15 Feb 2017 - 15:43 LauraBevilacqua ROC curves
PNGpng ImageOverlapDiagram1.png manage 37.4 K 06 Feb 2017 - 20:48 AaronAndrews Non-aligned image diagram
PNGpng Matched5stars.png manage 409.0 K 07 Feb 2017 - 15:19 AnishaVadher Stars matched up with accurate pixel coordinates
PNGpng PixelOverlapDiagram.png manage 34.7 K 06 Feb 2017 - 20:49 AaronAndrews Drizzle algorithm - pixel overlap diagram
PNGpng RealDoubleHR.png manage 63.9 K 08 Feb 2017 - 10:15 AnishaVadher Real Double Cluster without brightest star data
PNGpng Transformation10stars.png manage 1217.2 K 06 Feb 2017 - 16:44 AnishaVadher Transformation applied to 10 stars
PNGpng Transformation5stars.png manage 1173.7 K 07 Feb 2017 - 15:08 AnishaVadher Transformation matrix applied to 5 stars in frame
PNGpng purity_efficiency_eqs.png manage 5.1 K 07 Feb 2017 - 20:44 LauraBevilacqua p and e equations

Physics WebpagesRHUL WebpagesCampus Connect • Royal Holloway, University of London, Egham, Surrey TW20 0EX; Tel/Fax +44 (0)1784 434455/437520

Topic revision: r9 - 15 Feb 2017 - LauraBevilacqua

 
This site is powered by the TWiki collaboration platformCopyright © 2008-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding RHUL Physics Department TWiki? Send feedback