CUBE (Combined Uncertainty and Bathymetry Estimator)

CUBE (Combined Uncertainty and Bathymetry Estimator) is a method for automatically cleaning MBES data (10's to 100's of times faster than the standard processing approaches) and is very fast and statistically robust.

The technique is an error-model based, DTM generator that estimates the depth plus a confidence interval directly on each node point of a bathymetric grid.

The CUBE attributes each sounding with an estimate of horizontal and vertical error, uses a model of information propagation to transfer information about the depth from each sounding to its local neighborhood and then the grid nodes form depth hypothes.

The mainly questions are:

‘‘what is the true depth at certain point, given that all measurements have errors in all three dimensions?’’

‘‘How sure are we of that estimate?’’

The depth uncertainty that each sounding has increase with the TPU values and the horizontal distance from nodes. The point at which a sounding will no longer contribute to a node is dependent upon the IHO Order selected (the stricter the order chosen, the smaller radius of influence of that sounding) and the limit where the vertical uncertainty does not exceed the acceptable maximum according to the selected IHO order.

The CUBE allows multiple depth estimates/hypotheses to exist at each grid node and then uses “disambiguation” rules, typically based on a measure of local context, to determine which hypothesis at each node is the most “correct”.

The CUBE surface contains many layers (e.g. Deep, Density, Mean, Hypothesis Count etc.), one example is the Uncertainty layer, that shows the vertical uncertainty associated with the selected hypothesis. Lower uncertainties are better.