Floodfill algorithm improvements

  • To improve our algorithm we defined a star to be anything above 5 pixels in size.
  • It seems that all hot pixels were removed:
Floodfill_original.png

floodfill_imrpoved.png

  • Found an interesting structure:
interesting_stars.png
  • To improve center founding, we found x and y weighted averages, such that:
 \begin{equation} x_{\rm center} = \frac{\sum_{i = 1}^N x_i p_i}{\sum_{i=1}^N p_i}\end{equation}  \begin{equation} y_{\rm center} = \frac{\sum_{i = 1}^N y_i p_i}{\sum_{i=1}^N p_i}\end{equation}
  • where $x_{\rm center}$ and $y_{\rm center}$ are the x,y centers and $x_i$ and $y_i$ are the x and y coordinates of the individual pixels of the star and $p_i$ is the pixel value of the pixel, and $N$ is the total number of pixels of a given star.
  • An example of a found star center is shown below:
center_improved.png
  • The standard deviations were found using the following equations:
 \begin{equation} \sigma_x = \sqrt{\frac{\sum_{i = 1}^N (x_i - x_{\rm center})^2 p_i}{\sum_{i=1}^N p_i}}\end{equation}  \begin{equation} \sigma_y = \sqrt{\frac{\sum_{i = 1}^N (y_i - y_{\rm center})^2p_i}{\sum_{i=1}^N p_i}}\end{equation}  <br />

  • where $\sigma_x$ and $\sigma_y$ are the standard deviations in x and y.
  • Some information about found sigmas for all stars in this image:
  Minimum value Maximum Value Average Value
$\sigma_x$ 0.58 4.96 2.42
$\sigma_y$ 0.44 4.34 2.24

2D Fourier Transforms

  • Fourier transforms of such a star field will give regions in k space for both background and stars.
  • One of these can then be removed, leaving just stars or the background upon inverse Fourier transforming.
  • Initial attempts of this can be seen below.

  • This shows a location 1 image, with its background now clustered around the edges.
FT_Loc1_nocut.png
  • This shows an attempt at removing the background from the image.
IFT_Loc1_bkgcut.png
  • This shows an attempt at removing the stars from the image. Note the image is logged, in order to better see the remnants of the stars.

IFT_Loc1_starcut.png

Stacking

  • In order to reduce errors and create better contrast between stars and background, a preliminary stacking effort was performed for 6 images taken in location 1, the result of which can be seen below.
  • From this it appears that although some movement is expected across the images, their overlay does not exhibit any easily noticable affects caused by this. Note however no scientific analysis of the effects has been conducted thus far.Stacking_ap_6.png

-- ElenaCukanovaite - 01 Dec 2015

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng FT_Loc1_nocut.png r1 manage 1203.7 K 01 Dec 2015 - 16:33 DavidHadden  
PNGpng Floodfill_original.png r1 manage 87.8 K 01 Dec 2015 - 11:47 DavidHadden  
PNGpng IFT_Loc1_bkgcut.png r1 manage 49.6 K 01 Dec 2015 - 16:33 DavidHadden  
PNGpng IFT_Loc1_starcut.png r1 manage 1317.4 K 01 Dec 2015 - 16:33 DavidHadden  
PNGpng Stacking_ap_6.png r1 manage 42.2 K 01 Dec 2015 - 17:09 DavidHadden  
PNGpng center_improved.png r1 manage 68.5 K 01 Dec 2015 - 13:34 ElenaCukanovaite  
PNGpng floodfill_imrpoved.png r1 manage 37.8 K 01 Dec 2015 - 13:14 ElenaCukanovaite  
PNGpng interesting_stars.png r1 manage 62.2 K 01 Dec 2015 - 13:38 ElenaCukanovaite  
Edit | Attach | Watch | Print version | History: r9 < r8 < r7 < r6 < r5 | Backlinks | Raw View | Raw edit | More topic actions...

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

Topic revision: r7 - 01 Dec 2015 - DavidHadden

 
  • Edit
  • Attach
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2022 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