Data Reduction

Input data tables are limited in size. The maximum number of rows depends on the number of columns of the table. When an (imported) input table exceeds this number the possibility to reduce the number of rows is offered. The reduction is based on the Ramer-Douglas-Peucker algorithm.
Only tables which can logically be reduced can use this functionality. These are tables based on nodes/elements or distances. For instance the 'POLYDIF' table (IDENT based) or 'ADDIDENT' (X-P based, but idents cannot be interpolated) cannot be reduced, nor NODALL (node/distance based, but not logically to be interpolated).

The reduction procedure starts automatically when the table is tested, after single table import or when testing the table manually. Data is not reduced while importing via multi-table import. When the table is too large the message box below is presented:

ReduceConfirmation

Pressing 'No', the table is not reduced and the status of the table remains erroneous (see picture below), not allowing the function to be processed. At a later stage the table can be tested again, starting the reduction again.

ReduceNOT

Pressing 'Yes', the user is asked for the value of Ɛ, being the maximum distance between the original curve and the simplified curve. Press the 'Calculate' button (or press Enter) to start the reduction.

ReduceInput

When still too much rows remain it is reported as:

ReduceNotOK

and the 'Apply' button remains disabled.

When Ɛ is such that less than the maximal allowable rows remain it is reported as:

ReduceOK

Also a graph is available to see the difference between the original and reduced data set, for instance:

ReduceGraph

Having a large Ɛ, few rows remain and the graph would look like shown below.

ReduceGraphHighE

In this example, even when just 24 of the original 1155 rows remain the reduced curve is pretty similar to the original one.

Pressing the 'Apply' button the original data set is replaced by the reduced data set. Within one session the original data set can be restored using the 'Previous version' button Undo. When closing and re-opening this project the original data set will be lost, but can be imported again of course.

 


Data reduction, last changed:14/09/2016