RIET USERS: Background modelling

Lachlan Cranswick ( lachlan@melbpc.org.au )
Tue, 14 Oct 1997 13:41:17 +1000

Sent on behalf of Monica Rubio of the University of Santiago de Cuba
E-mail: MRubio@cnm.uo.edu.cu. If any responses of interest, please
consider Cc'ing the list.
Cheers, Lachlan.

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
From: Monica Rubio <MRubio@cnm.uo.edu.cu>
Subject: Background modelling

I have been trying to model the background of a step-scanned x-ray pattern
with no satisfactory result. After determining the peak positions using a
search procedure, I have the following unsolved questions:

1. How to discriminate the background points from the peak points?

2. What are the best functions for modelling the background? (Considering
different types of backgrounds: nearly flat, with broad humps in the
background curve, with background level initially high and then falling
down smoothly, etc.).

3. How to refine the parameters of the function(s) selected?

4. Need(?) of the use of a scale factor.

Any help to solve these questions will be appreciated as I have no experience
nor literature on the subject.

Monica Rubio
Computer Science Dept., Fac. of Sciences
University of Oriente
Santiago de Cuba 90500, Cuba
email: MRubio@cnm.uo.edu.cu