Home Page       Mon Dec 1 02:11:21 JST 1997

Linearity between magnitude and pixel value

Then I confirmed that stars of same brightness look same wherever they locate on the image. In addition, restoration of level correction in the method mentioned in this article causes not only dissolution of the difference depending on the position but also restoration of the linearity between magnitude and pixel value.

I used the image of Nova Cas 1995 for this experiment. This is originally a printed picture and scanned.

Nova Cas 1995
Photo: Seiichi Yoshida
Jan. 9, 1996, 19:01:30 JST (1 min)
Fujishiro Town, Ibaraki Pref. Japan
25-cm f/6.3 Schmidt-Cassegrain
Konica color GX3200

Let's pick up A C in circumference and D F in center. The table below shows the difference of pixel value amount of the detected stars in the old method described in [2] and in the new method mentioned in this article, and the true magnitude of the stars in the Tycho Catalog.

The mean magnitude of A C is 9.25 mag and D F is 10.83 mag. The ratio in brightness is 4.25 times. In the old method, the ratio of pixel value is only 2.1, probably because A C in circumference look fainter. In the new method, the ratio is 4.4 and very similar to the ratio of magnitude, which means the linearity is restored.