_______ SaGA Gallery, part 2 _______

Here are a few pictures illustrating some concepts and algorithms used in SaGA:

Construction of a convex hull - a picture illustrating a method of producing multi-dimensional convex hull of a set of points

Empty circumference property Illustration of the empty circumference property of a Delaunay triangulation: there are no other points inside a circumference of each triangle.


Delaunay triangulation as a convex hull on a sphere This figure illustates the method of constructing Delaunay tesselation in dimension k: map dataset on a sphere in k+1 dimension, construct a convex hull of this mapped set and you get a Delaunay of original set


Interpolation by triangulation This figure shows examples of interpolation from irregular points by triangulation method. It shows gridding results from the same set with 3 different options of function INTERPTR: with pure linear triangular interpolation, with extrapolation beyond the convex hull (with function EXTRAPTR) and blending with gradient information (with function GRAD2EST).


Interpolation by objective mapping This figure shows gridded surface and associated error map of noisy irregular set of points, obtained with OBJMAP function.


Filling missing values Restoration of missing values of a matrix with function FILLMISS. Missing values are adaptively interpolated from available neighbouring elements.

Quadtree division Illustration of the Quadtree concept: To obtain values in a certain block only points within this block and its neighbours are used; if some neighbouring blocks are "underpopulated" their own "secondary" neighbours are also included.


And here is the function sagapic.m which can produce all these pictures. It can be used as a demo for many programs in the SaGA Toolbox.


Back to the SaGA Homepage