File Exchange

image thumbnail

Fast Guassian Blur

version (810 KB) by Royi Avital
Fastest Implementation of Gaussian Blur


Updated 22 Jun 2018

GitHub view license on GitHub

Evaluation of few methods to apply Gaussian Blur on an Image.
The implemented methods:
1. Truncated Gaussian Kernel (FIR Filter).
2. Gaussian Blur approximation using Box Blur (Central Limit Theorem).
3. Gaussian Kernel Approximation using IIR Filter.

Comments and Ratings (3)

Thank you for sharing this. It is a very nice comparison of different ways to implement the Gaussian blur. Maybe in the overview you could also share the performance comparison results. Also note that if you have the image processing toolbox, then imgaussfilt(...,'FilterDomain','spatial') will give you a truncated gaussian 2D FIR filter.



Updated Description.

MATLAB Release Compatibility
Created with R2014a
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired by: FilterM

Inspired: Blind Gaussian De-blurring of an Image