Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.sao.ru/precise/Midas_doc/doc/94NOV/vol2/node127.html
Дата изменения: Fri Feb 23 14:01:40 1996 Дата индексирования: Tue Oct 2 18:27:21 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п |
This problem is examined in Buonanno and Iannicola, 1989.
In short, ROMAFOT performs the best fit by comparing a given pixel
to the value of the PSF at the centre of the pixel. This approximation
is valid if the pixel size is small compared with the scale length of
the point images. If R is the ratio between the FWHM of the PSF
and the pixel size, then a value of is the limit of validity
of the quoted approximation and, correspondingly, the limit beyond
which we enter in the regime of big pixels. With the commands
below the integral of the PSF over the pixel area is computed
via the Gauss - Legendre integration formulae which are characterized
by a high precision for a correspondingly small number of subpixels
where the PSF must be calculated.
Since the number of subpixels depends on the intensity of the star, on the local gradient of the PSF and so on, two commands are used. The first command prepares a file where the subpixel values are computed according to the different parameters of the problem (aperture of the telescope, exposure time, seeing, required accuracy, magnitude of the star, distance of the pixel from the centre of the star and so forth). The second command performs the non - linear fitting by comparing the integral of the PSF over the subpixel values.