1. Home

Absolute Integrator

Absolute Integrator provides a set of tools for the automated quantitative analysis of atomic-resolution ADF images.

There are various methods for achieving this given in the current literature and it is designed so that the user can choose more/fewer/different module so as to reproduce the different approaches.

The current core of the method relies on four stages:

1. Image pre-processing (see below),
2. Image-feature identification (usually peak-finding),
3. Normalisation to the incident beam-current, and
4. Integration at each atomic column.

For each of these three stages there are several choices. In the repository folders exist for each block and scripts within these for each method choice. These include:

Image Pre-processing:

• Magnification calibration (required),
• Background subtraction (optional).

Feature Identification (choose one):

• Peak-finding by the Ranger method,
• Peak-finding by template matching,
• Peak-finding by basis rectors.

Image Normalisation (choose one):

• Detector sensitivity profile (DSP) method,
• Electron flux weighted (EFW) method [ADD REFERENCE HERE].

Column Integration (choose one):

• Voroinoi cell-wise integration,
• Multiple overlapping Gaussians method.

It is envisaged that more methods may be added in the future so all should be developed to use a standardised format for the parsing of the key parameters from one to the other.