Home Preprocessing Arithmetic Conversions  
See also: Mathematical Transformations, Scaling of the Data, PerformBulkTransform


Predefined Arithmetic Conversions 

ImageLab provides a set of predefined, frequently used mathematical conversions. Some of the conversions include a constant parameter A which can be set by the user. The data to be processed can be restricted in two ways: (1) the user can select a mask which excludes the masked pixels from processing, and (2) the range of layers to be processed can be constrained.
When applying an arithmetic conversion some of the resulting values may be indefinite. ImageLab distinguishes between "hard errors" and "soft errors". Hard errors can be recognized before the transformation starts. Those errors are reported immediately and the conversion cannot be started. Soft errors emerge during the actual calculations. They are counted and reported to the user at the end of the conversion process. Each cell of the data cube which gives rise to a soft error is filled whenever possible by a reasonable value (which results from the consideration of the mathematical limit of the particular transformation) or else by the indefinite substitute value (which is, by default, set to 1, but can be adjusted in the preferences dialog). The following table provides a list of available conversions (x represents the spectral data):
The following diagram shows the transfer characteristic of the sigmoid function for an amplitude of 1.0 and various slopes:
