Meade Image Processing Bedienungsanleitung PDF herunterladen (Seite 17)

Optional mathematical treatment of dark subtraction and flat fielding

1. The reading on each pixel is N = D + S × L, with dark current D, sensitivity S, and

light L. Both D and S fluctuate from pixel to pixel.

2. A dark frame has no light, L = 0, so N

darkframe

= D.

3. A flat frame is evenly illuminated, L = 1, hence N

flatframe

= S.

4. Now to restore the reading due to true light only in the image, we must first dark

subtract, to find N

dark subtracted

= N - N

darkframe

= (D + S × L) – D = S × L ; then flat field (i.e.

divide by a flat frame), to find N

ds’d+flatfileded

= N

dark subtracted

/ N

flatframe

= (S × L) / S = L.

5. If we do it in the wrong order, flat field first, then dark subtract, we find the wrong

pixel reading N / N

flatframe

- N

darkframe

= (D + S ×L) / S –D = L + (1/S-1) × D ≠ L; some of the

dark current remains on the picture, together with its noisy fluctuation.

This discussion also illuminates the importance of the linearity of the CCD chip. When

the chip is approaching saturation, in our camera at S × L > 40,000, the sensitivity in fact

declines and the reading will be less than S × L would indicate. Brighter parts of the images

will not be as bright as they should be; contrast is lost. Over S × L > 65,535 the pixel

reading stops growing altogether and the image becomes completely clipped.

1 2 ... 12 13 14 15 16 17 18 19 20 21 22 ... 28 29

Keine Kommentare

Meade Image Processing Bedienungsanleitung Seite 17