What is Gaussian blur in image processing?
Gaussian distribution
In image processing, an image is processed with the help of the function Gaussian distribution to reduce the noise or to simplify details.
The Gaussian distribution (also known as the normal distribution) is a bell-shaped curve with an equal number of observations above and below the mean value. It is necessary to grasp the definitions of "mean," "median," and "mode" to comprehend Gaussian distribution.
The "mean" is the computed average of all values.
The "median" is the value in the distribution's center (midpoint).
The "mode" is the value seen most frequently during the measurement.
Gaussian formula
The Gaussian distribution in 1-D is:
Where
Gaussian noise
The main causes of Gaussian noise in digital images occur during captures, such as sensor noise induced by inadequate lighting, excessive temperature, or transmission. When smoothing an image, an undesired effect may arise in blurring fine-scaled image borders and features since they correlate to blocked high frequencies. Mean (convolution), median, and Gaussian smoothing are traditional spatial filtering techniques for noise reduction.
Â
Gaussian filter
The Gaussian filter blurs the desired area and cuts the noise with higher frequencies. It works the same as mean filters while representing average weight uniformly. These are linear filters that reduce the noise and blur the edges effectively. They are created as matrices in digital image processing, passing through each pixel of the selected portion.
Â
Use cases
A few use cases of Gaussian image processing are:
It helps to eliminate noise from blurry images.
It is used to reduce noise from the image taken in low light.
It can eliminate the bright pixels.
It helps to make edges smooth.
It is rationally symmetric.
We can define the degree of smoothing.
Free Resources
