next up previous
Next: Removing image contamination
Up: Deriving accurate background values
Previous: Global background value determination

Local background value determination

 

If you are intending to determine profiles for a large number of galaxies on an image and the image is not perfectly flatfielded then the background values in the image region surrounding each of the galaxies may be determined using LOBACK.

The first requirement for this is a text file containing a list of the coordinates of the objects on the image. This may be derived in a number of ways. The most obvious is to use the KAPPA application CURSOR (with LOGFILE assigned a value). This method is convenient when only a small number of galaxies are required from each image. However, if all the galaxies/objects on an image are to be examined then the text files generated by PISA, RGASP's IMAGES or IRAF's FOCAS might form the the basis of an input file to LOBACK.

The input file for LOBACK should (in the simplest case) contain two columns, they represent the x and y co-ordinates respectively in the Current co-ordinate system of the NDF. LOBACK then determines modal pixel values, pixel value standard deviation and background standard deviation values at each of the image locations described in the text file. Optionally (parameter THIRD=FALSE), you may define a third column representing the minimum number of pixels to be used in determining the modal pixel value or alternatively (parameter THIRD=TRUE) the number of pixels believed to be in the object found at that location. In the latter case the number of pixels used, and the size of the image area they are taken from is adjusted to ensure the sample contains a significant number of non-object pixels - hopefully sky. In either case, the software imposes a lower limit on the number of pixels to be used, thereby reducing the effect of a sparse pixel value histogram.

The application can be used with the following syntax:

% loback in=p2 infile=coords.dat sfact=2 third=true out=backs.dat width=64
This reads the galaxy co-ordinates from a file called coords.dat. The information in the third column of the text file is assumed to be the number of contiguous pixels found in the galaxy by PISA, IMAGES or FOCAS. The pixels used to make up the histogram required will be taken from an area of the image 64x64 pixels in size. In the unlikely event of an image being smaller than the requested number of pixels required, all the non-bad image pixels will be employed. Other examples are given in Appendix B.1

If an object location is not within the bounds of the image requested then an error message is generated for that object, but the program continues reading the object list working on each in turn. The modal pixel values generated use the same methods as HISTPEAK, i.e. (un)smoothed histogram, projection and interpolation (see HISTPEAK).

An example output file is shown in Appendix F and described in Section 15.



next up previous
Next: Removing image contamination
Up: Deriving accurate background values
Previous: Global background value determination

ESP --- Extended Surface Photometry
Starlink User Note 180
Norman Gray
Mark Taylor
Grant Privett
27 January 2000
E-mail:ussc@star.rl.ac.uk

Copyright © 2002 Council for the Central Laboratory of the Research Councils