Algorithms

• QuickSampling (QS) (aka. Quantile Sampling) is a general-purpose pixel-based MPS algorithm that is designed to be robust, efficient, and run in constant time. QS was designed to adapt to your problem; it can be used to do (un)conditional simulation, gap filling, or even downscaling, using continuous or categorical variables or a combination of both. The code was developed without restrictions regarding the dimensionality of the data (e.g. 1D, 2D, 3D, nD).
• Anchor Sampling (AS) is a location-anchored variant of QS for aligned training-image stacks. It preserves the same path-based simulation philosophy and neighborhood matching logic, but the admissible sampled center value is restricted to the same spatial coordinate across the TI stack. Pronounce it like the French as 🂡, or letter by letter as A, S.
• Narrow Distribution Selection (NDS) is an algorithm specifically targeted to simulate spectrally enhanced remote-sensed imagery. It requires an external variable (for example, a grayscale image) to control the simulation (of colors).
• SNESIM is a categorical search-tree-based MPS algorithm with a dedicated multigrid workflow.
• AutoQS is a calibration workflow used to explore QuickSampling settings.

The framework can be easily extended to handle most codes that use gridded data. Currently, any compiled code or Python code can be handled.