next up previous contents index
Next: F.5 Assessment of the Up: F. Optimising ISOCAM Data Previous: F.3 Variable Flat-Field


F.4 Bad Pixel Identification

Usual deglitching methods that work on the time history of a given pixel are good to remove extremely deviant flux values. We can go further in the reduction of the noise level by working sky position by sky position instead of working on the time history of every pixel. For a given sky position ($\alpha$, $\delta $), we compare the $N$ pixels in the data cube that have seen that position with the average flux in the map at the vicinity of the position ($\alpha$, $\delta $).

Three cases are possible, and here is how we deal with each case:

  1. Most of the $N$ pixels are around the local average $B$.
    The new sky flux is the average of ( $B - 3 \times noise < I_{obs} < B + 3 \times noise$)
  2. Most of the $N$ pixels are above the local average B.
    The new sky flux is the average of $I_{obs} > B$
    This case corresponds to point sources.
  3. Most of the $N$ pixels are under the local average B.
    The new sky flux is the average of ( $B - 3 \times noise < I_{obs} < B$).
This method allows us to reduce the noise level and to keep a good photometry of point sources. This is the final step of the data processing. The sky images obtained after bad pixel removal of the first and second GRB observations are presented in Figures F.3d and  F.3e.

next up previous contents index
Next: F.5 Assessment of the Up: F. Optimising ISOCAM Data Previous: F.3 Variable Flat-Field
ISO Handbook Volume II (CAM), Version 2.0, SAI/1999-057/Dc