Automatic samples identification

straditize has reader specific sample identification algorithms implemented. The overall algorithm is based on two steps:

1. For each column: Identify the intervals that contain exactly one samples (the rough locations)

2. Align the overlapping intervals between the columns to estimate the exact location

Where step 2 is the same for all diagram types, step one is diagram specific. For bar diagrams, we use the bars that have been identified during the digitization (see Digitizing bars).

For area and line diagrams on the other hand, we identify the local extrema (minima and maxima) for each column. This strategy works well for pollen diagrams since they are supposed to sum up to 100%. However, this does not generally work for all stratigraphic diagrams.