-- LauraBevilacqua - 18 Oct 2016

To do by Friday:

  • Select star cluster for observations
Anisha:
  • Fix histrogram from last week
  • Plot median as a function of exposure time
  • Plot error as function of exposure time
Aaron:
  • Figure out how to save simulated images as FITS files.
  • Give Laura test fits file, with recorded star positions and intensities/counts.
Laura:
  • Convert Aaron's fits file to image
  • Test whether program counts accurately
Weekly Progress:

Anisha:

  • Corrected histrogram of number of photoelectrons for a 0.1s exposure in the dark frame with 500 bins - logarithmicly normalized:
    Histogram_of_number_of_pixel_values_for_0.1s_exposure_500_bins.png

  • The median intensity as a function of exposure time.:
    The_median_intensity_as_a_function_of_exposure_time_with_best_fit.png

  • Line of best fit: y = 917.43 + 0.08x

  • Error as a function of exposure time:
    Error_as_a_function_of_exposure_time_with_best_fit.png

  • Line of best fit: y = 15.76 + 0.73x
 Mean number of effective photoelectrons : \begin{equation} \nu = \nu_0 + \mu t. \end{equation} Actual number of photoelectrons: \begin{equation} n = n_0 + m, \end{equation} where $n_0$ follows a Gaussian distribution and may come from electronic noise and $m$ follows a Poisson distribution as a function of $\mu t$. Error in $m \sigma_m$ come from $\sqrt{\mu t}$. Therefore the intercept of the $n_{peak}$ (the median) as a function of exposure time is $\mu$ and the intercept is $\nu_0$.

Aaron:

  • Implemented a true 2D Gaussian with correlation between x and y (with correlation coefficient P (rho)):
 \begin{equation} f\left ( x,y;x_0 ,y_0 ,\sigma _{x} ,\sigma _{y} ,\rho \right )= \frac{1}{2 \pi \sigma _{x} \sigma _{y} \sqrt{1-\rho ^{2}}} \exp {\left [ - \frac{1}{2\left ( 1-\rho ^{2} \right )}\left ( \left (\frac{ x-x_0 }{2{\sigma _{x}}^{2}} \right )^2 + \left (\frac{ y-y_0 }{2{\sigma _{y}}^{2}}\right )^2 -2\rho \left (\frac{ x-x_0 }{2{\sigma _{x}}^{2}} \right )\left (\frac{ y-y_0 }{2{\sigma _{y}}^{2}}\right ) \right )\right ]}  \end{equation}
  • (Note: normalisation factor is ignored so far for programming purposes; maximum pixel value is being used in its place)
  • Integration of 2D Gaussian in this form is done numerically.
  • Tested for P =0, P =0.5, P =0.999 (sigma_x=sigma_y=3 in each case)
  • SkewTest2.png
  • (Note - image is rotated 90 degrees clockwise)
  • Also - found a way to save images as FITS files, similar to images gained from observations.
  • Simulated test images for Laura - all on 100x100 pixel grids, using calculated mean pixel values instead of randomised poisson values:
    • 1 star at (50,50), peak=60000.
    • FinderTest1a.png
    • 3 stars with decreasing intensities: (25.25), peak=10000; (50,50), peak=1000; (25,25), peak=100
    • FinderTest2.png
    • 8 stars with peak=60000, in pairs which are closely spaced: (45,20), (55,20); (44,40), (56,40); (43,60), (57,60); (42,80), (58,80)
    • FinderTest3.png
    • 8 stars with peak=60000, peaks are out of frame: (100,20), (102,40), (104,60), (106,80), (-8,20), (-10,40), (-12,60), (-14,80)
    • FinderTest4.png
Laura:
  • Fits files from Aaron successfully converted to monochrome images
  • Images scanned for stars and correct star count is returned
  • Edge detection helps identify star position

  • The maximum pixel value was found to be 65195 from the fits file below, as opposed to Aaron's given value of 65535
    FinderTest1a.jpg
  • Threshold of 100 set so if pixels are above this value they are set as white background, if below they are coloured black (remember reverse grayscale):
    FinderTest1a_mon.jpg
  • Edge detection:
    FinderTest1a_count.png

  • Maximum pixel value found to be 62673, compared to Aaron's 60000:
    FinderTest3.jpg
  • Threshold set as 20 so that stars don't touch, achieves accurate count of 8:
    FinderTest3_mon.jpg
    FinderTest3_count.png
Topic attachments
I Attachment History Action Size Date Who CommentSorted ascending
PNGpng Histogram_of_number_of_pixel_values_for_0.1s_exposure_500_bins.png r1 manage 18.4 K 20 Oct 2016 - 17:53 AnishaVadher Corrected histrogram of number of photoelectrons for a 0.1s exposure in the dark frame with 500 bins - logarithmicly normalized
PNGpng Error_as_a_function_of_exposure_time_with_best_fit.png r1 manage 31.3 K 20 Oct 2016 - 18:01 AnishaVadher Error as a function of exposure time
PNGpng FinderTest1a.png r1 manage 14.5 K 21 Oct 2016 - 12:16 AaronAndrews Finder test 1: 1 star
PNGpng FinderTest2.png r1 manage 16.5 K 21 Oct 2016 - 12:17 AaronAndrews Finder test 2: 3 stars, decreasing intensities
PNGpng FinderTest3.png r1 manage 35.9 K 21 Oct 2016 - 12:17 AaronAndrews Finder test 3: 8 stars, closely spaced
PNGpng FinderTest4.png r1 manage 17.0 K 21 Oct 2016 - 12:18 AaronAndrews Finder test 4: 8 stars, out of frame
JPEGjpg FinderTest1a.jpg r1 manage 6.5 K 20 Oct 2016 - 18:07 AnishaVadher  
PNGpng FinderTest1a_count.png r1 manage 12.8 K 20 Oct 2016 - 18:07 AnishaVadher  
JPEGjpg FinderTest1a_mon.jpg r1 manage 5.4 K 20 Oct 2016 - 18:07 AnishaVadher  
JPEGjpg FinderTest3.jpg r1 manage 14.1 K 20 Oct 2016 - 18:07 AnishaVadher  
PNGpng FinderTest3_count.png r1 manage 15.5 K 20 Oct 2016 - 18:07 AnishaVadher  
JPEGjpg FinderTest3_mon.jpg r1 manage 6.2 K 20 Oct 2016 - 18:07 AnishaVadher  
PNGpng SkewTest2.png r1 manage 18.6 K 18 Oct 2016 - 22:46 AaronAndrews Testing x and y correlation (P) for P=0, P=0.5, P=0.999
PNGpng The_median_intensity_as_a_function_of_exposure_time_with_best_fit.png r1 manage 36.2 K 20 Oct 2016 - 17:55 AnishaVadher The median intensity as a function of exposure time.
Edit | Attach | Watch | Print version | History: r11 < r10 < r9 < r8 < r7 | 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: r11 - 21 Oct 2016 - AaronAndrews

 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2021 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