Enum Interpolation2DType

Namespace

AiDotNet.Enums

Assembly

AiDotNet.dll

Specifies different methods for interpolating 2D data points to create a continuous surface.

public enum Interpolation2DType

Fields

Bicubic = 1

A smoother interpolation method that uses cubic polynomials in both x and y directions.

For Beginners: Bicubic interpolation creates a smoother surface than bilinear by using curved lines instead of straight lines between points.

Think of it as:

• Like bilinear, but with curves that create smoother transitions
• Similar to how high-quality image resizing works
• Preserves smoothness better than bilinear
• Still relatively fast, but more accurate

Best used when:

• You need smoother results than bilinear provides
• Working with data that should be continuous and smooth
• Resizing images or other grid data with better quality
• You want a good balance between speed and smoothness

Bilinear = 0

A simple, fast interpolation method that uses linear interpolation in both x and y directions.

For Beginners: Bilinear interpolation is like drawing straight lines between your known points and using those lines to estimate values in between.

Think of it as:

• The simplest and fastest method
• Like stretching a flat sheet over your data points
• Good for when you need quick results and don't need perfect smoothness
• Similar to how a low-resolution image looks when zoomed in

Best used when:

• Speed is more important than accuracy
• Your data is already fairly smooth
• You're working with grid-like data (like images)
• You need a simple, predictable result

CubicConvolution = 7

An interpolation method that preserves the sharpness of edges while providing smooth results elsewhere.

For Beginners: Cubic Convolution is similar to bicubic interpolation but uses a wider range of neighboring points to calculate each value.

Think of it as:

• An enhanced version of bicubic interpolation
• Better at preserving edges and details
• Creates smooth results without excessive blurring
• Commonly used in image processing and remote sensing

Best used when:

• You need to preserve edges and details
• Working with images or grid data
• You want better quality than bicubic but without artifacts
• Your data contains both smooth regions and sharp transitions

Kriging = 3

A geostatistical method that uses spatial correlation between data points.

For Beginners: Kriging is an advanced method that considers how data points relate to each other based on distance and direction. It was originally developed for mining and geology applications.

Think of it as:

• A "smart" method that learns patterns from your data
• Takes into account how values tend to vary with distance
• Provides both estimated values AND uncertainty estimates
• More computationally intensive but potentially more accurate

Best used when:

• Your data has spatial patterns or trends
• You need to know how confident you can be in the interpolated values
• Working with geospatial data like elevation, rainfall, or mineral concentrations
• You have enough data points to establish reliable spatial relationships

MovingLeastSquares = 5

A flexible method that fits local polynomial functions to nearby data points.

For Beginners: Moving Least Squares creates a smooth surface by fitting small, simple mathematical functions to groups of nearby points.

Think of it as:

• Like having many small "patches" that blend together
• Each area is influenced mainly by nearby points
• Creates a smooth surface that adapts to local patterns
• More flexible than simpler methods

Best used when:

• Your data has different patterns in different regions
• You need a smooth result that adapts to local features
• Working with complex surfaces like terrain or 3D models
• You need better quality than simple methods but don't want the complexity of Kriging

MultiQuadratic = 6

An interpolation method using radial basis functions with multiquadratic form.

For Beginners: MultiQuadratic interpolation uses special mathematical functions that create smooth hills and valleys centered at each data point.

Think of it as:

• Placing a smooth bump or dip at each data point
• These bumps blend together to form a continuous surface
• Creates very smooth results even with scattered data
• Good for capturing both local and global patterns

Best used when:

• You need very smooth interpolation
• Working with scattered data points
• Your data represents a physical phenomenon that should be smooth
• You need accurate results and smoothness is important

ShepardsMethod = 4

A distance-weighted interpolation method that gives more influence to nearby points.

For Beginners: Shepard's Method calculates values based on the idea that nearby points should have more influence than distant points.

Think of it as:

• Like a weighted average where closer points count more
• The influence of each point decreases with distance
• Simple to understand and implement
• Works with irregularly spaced data points

Best used when:

• You have scattered data points (not in a grid)
• Closer points should logically have more influence
• You need a method that's intuitive and relatively simple
• You want to avoid complex mathematical models

ThinPlateSpline = 2

A flexible interpolation method that minimizes the bending energy of a thin metal plate.

For Beginners: Thin Plate Spline interpolation creates a smooth surface that passes through all your data points while minimizing the overall "bending" of the surface.

Think of it as:

• Like placing a thin, flexible metal sheet over your data points
• The sheet bends to touch all points but stays as flat as possible elsewhere
• Creates very natural-looking smooth surfaces
• Good for scattered (non-grid) data points

Best used when:

• Your data points aren't arranged in a grid
• You need a smooth surface that passes exactly through your data points
• You're working with geographic or spatial data
• You want natural-looking results for physical phenomena

Remarks

For Beginners: Interpolation is like "filling in the blanks" between known data points. Imagine you have temperature readings from several weather stations across a city, and you want to estimate the temperature at locations between these stations. Interpolation methods are different mathematical techniques to make these estimates.

Each method has different strengths:

• Some are faster but less accurate
• Some preserve certain properties of your data better than others
• Some work better for smooth data, others for data with sharp changes

The right choice depends on your specific data and what properties you want to preserve.