next up previous contents
Next: 4.2 Microphonic Noise Detection Up: 4 Raw Image Screening Previous: 4 Raw Image Screening

4.1 Bright-Spot Detection

Long IUE exposures characteristically contain ``bright spots'', i.e., pixels with unusually high data number (DN) values which are comprised of discrete impulse noise often reaching the saturation level. Such bright spots are thought to be caused either by permanent blemishes in the target surface, by extraordinarily sensitive (``hot'') pixels which result in recurrent bright spots at fixed locations, or by radiation-induced events within the UV converter which result in randomly placed, nonrecurrent bright spots (Ponz, 1980).

Ponz (1980) has described an algorithm for detecting in raw images bright spots of either kind on the basis of their limited spatial extent and unusual brightness values, primarily through a median filtering technique. The NEWSIPS bright-spot detection algorithm is based on the procedure used in standard IUESIPS processing, which incorporates this method to flag bright spots as described below.

Let DN (i,j) be the DN value of the pixel at line i, sample j. Further, let AVE and MED represent operators which return the weighted average and median values of their argument, respectively. Then the pixel at (i,j) is detected as a bright spot if:

\begin{displaymath}
DN(i,j) \gt AVE\{DN(k,l)\} + \Delta~~~~{\rm and}~~~~DN(i,j) \gt
MED\{DN(k,l)\} + \Delta \end{displaymath}

where $\Delta$ is a DN threshold value, and (k,l) are positional elements of a 7-pixel spatial window centered on the pixel at (i,j) and oriented on a diagonal (i.e., nearly along the dispersion direction). The condition in the first equation is included to reduce the number of times the median operation in the second equation is performed.

In practice, the spatial windows are weighted according to the weights (0, 0, 1, 0, 1, 0, 0), and a threshold value of $\Delta$ = 90 DN is employed. The area of the image searched for bright spots corresponds to the entire camera faceplate regardless of dispersion. This differs from the IUESIPS approach, which only examined the regions containing spectral information. Pixel locations detected as bright spots are written to an output flag file (Chapter 3) subsequently read by the spectral extraction routines ( Chapters 9  and  10) so that extracted fluxes derived from bright spot pixels may be flagged appropriately.

Several reports have been written that list the permanent or recurrent bright spots for the three cameras. These bright spots are not automatically flagged like the permanent ITF artifacts (see Chapter 6.4.3.1 ), but rather only if they trigger the bright-spot detection algorithm. Ponz (1980) has published partial listings of recurrent bright spots in the LWR and SWP cameras which are listed in Tables 4.1 and 4.2. The table entries include the line and sample positions in the raw frame of reference and the approximate corresponding wavelengths for the various dispersion modes and apertures. The ``B'' notation means the background spectrum is affected. This work has been supplemented by Imhoff (1984a), who provided positions of additional permanent blemishes in the LWP and LWR cameras, given in Table 4.3.

   
Table 4.1:  Hot Pixels in the LWR Camera (Ponz,1980)
Raw Image Low Dispersion (Å) High Dispersion (Å)
Line Sample Lg. Ap. Sm. Ap. Lg. Ap. Sm. Ap.
126 291     1919.3 1904.8 B/1920.5
170 200 1780 1775 B    
175 369     2172.5 2153.6 B/2173.9
178 610     2732.0 2733.8
208 391     2258.5 B/2280.0 2282.4 B
215 326   2130 2135.3 2117.0 B/2136.7
257 323 2190   2198.2 2199.7 B/2178.8
333 317     2288.9 2290.3/2268.0 B
412 385     2570.2 2543.8 B/2572.0 B
434 479     2818.7 2786.3 B/2820.5 B
518 545     3084.0 3086.0
532 307     2550.8/2579.2 B 2552.3
680 332     2838.0 2839.8

 
 
Table 4.2:  Hot Pixels in the SWP Camera (Ponz, 1980)
Raw Image Low Dispersion (Å) High Dispersion (Å)
Line Sample Lg. Ap. Sm. Ap. Lg. Ap. Sm. Ap.
292 413     1379.6 B/1393.6 1378.7 B/1392.6
352 501     1330.2 B/1343.0 1342.2
392 127 1795 B   1859.1 1857.8
398 521     1357.9 B/1371.4 1357.0 B/1370.4
410 535     1358.5/1372.0 B 1357.6/1371.0 B
482 342     1686.7 1685.6
568 127     2060.2 2058.9
611 387     1779.0 B/1756.5 B 1778.0 B/1755.3 B

Several publications by Crenshaw et al. detail the recurrent bright spots, which are labeled ``camera artifacts'', in low-dispersion (Crenshaw et al. 1990) and high-dispersion (Crenshaw et al. 1996) spectra. These features appear at fixed locations in spectra with long exposure times yet are not detectable in spectra with exposure times shorter than an hour. They also determined that the artifacts appear to scale in intensity with the total background exposure level. Artifact positions of the more prominent features are listed in Table 4.4 (low dispersion) and Table 4.5 (high dispersion) as a function of wavelength.
 
 
Table 4.3:  Permanent Blemishes in the LWP and LWR Cameras (Imhoff, 1984a)
  Raw Image  
Camera Line Sample Comments
LWP 101 525  
  205 319  
  396 384 Fuzzy Patch at $\lambda\sim$2482Å in order 93
  409 208 Hole at $\lambda\sim$2880Å in order 80
  426 435  
  455 35  
       
LWR 169 499  
  364 60  


(Crenshaw et al. 1990)
 
 
Table 4.4:  Low-Dispersion Camera Artifacts (Å)
  Source Type
Camera Point Extended
LWP None None
     
LWR 2256 2256
  3087 3087
     
SWP 1279 1279
  1288 1288
    1491
    1535
  1663  
    1750


(Crenshaw et al., 1996)
 
 
Table 4.5:  High-Dispersion Camera Artifacts
LWP LWR SWP
Order $\lambda$ Order $\lambda$ Order $\lambda$ Order $\lambda$
Number (Å) Number (Å) Number (Å) Number (Å)
93 2483.3 122 1888.3 119 1154.9 78 1757.7
80 2880.4 112 2067.3 119 1160.0 78 1762.4
    112 2074.4 118 1165.9 77 1790.1
    107 2167.1 118 1172.8 77 1795.5
    106 2171.7 116 1195.5 76 1810.1
    100 2303.2 116 1195.8 72 1914.3
    100 2322.3 115 1205.9 72 1916.0
    94 2450.1 108 1281.7 70 1975.0
    93 2501.1 106 1298.6 70 1980.6
    87 2670.1 104 1320.8 69 2005.0
    86 2694.4 103 1337.8 69 2009.5
    84 2756.2 101 1369.8 67 2059.1
    82 2808.4 95 1453.6 67 2059.5
    82 2829.0 93 1474.3    
    80 2900.6 93 1481.8    
    77 3010.7 93 1483.0    
    76 3029.6 93 1487.7    
    75 3066.4 92 1500.9    
    75 3083.1 92 1504.8    
    75 3100.0 92 1505.5    
    72 3194.3 90 1540.0    
    72 3195.2 89 1549.3    
        89 1552.0    
        88 1566.5    
        88 1573.0    
        87 1577.9    
        87 1583.9    
        87 1593.2    
        87 1593.5    
        86 1598.5    
        86 1603.6    
        82 1691.1    
        80 1725.9    
        79 1755.5    


next up previous contents
Next: 4.2 Microphonic Noise Detection Up: 4 Raw Image Screening Previous: 4 Raw Image Screening
Karen Levay
12/4/1997