Multivariant Kernel Regression and Smoothing

A multivariant Gaussian kernel regression and smoothing tool.

Description

This function implements multivariant Gaussian kernel regression and smoothing. A few examples show how it can be used to smooth 3-d noisy images with missing data.

For tutorial purpose, the function is implemented in the simplest form. Therefore, it is particularly suitable for beginners to learn kernel regression.

The efficiency is improved in the updated version by searching neighbours in the effective range.

Acknowledgements

The author wishes to acknowledge the following in the creation of this submission:
Kernel Smoothing Regression
This submission has inspired the following:
Efficient Kernel Smoothing Regression using KD-Tree

• MATLAB release: MATLAB 7.5 (R2007b)

[V. Poor]

 

[Turkay YILDIZ]

 

[Masoud]

as by default z=x. When mean(x)=0, the r.f will be NAN!