Exercise

# Smoothing

Smoothing can improve the **signal-to-noise ratio** of your image by blurring out small variations in intensity. The Gaussian filter is excellent for this: it is a circular (or spherical) smoothing kernel that weights nearby pixels higher than distant ones.

The width of the distribution is controlled by the `sigma`

argument, with higher values leading to larger smoothing effects.

For this exercise, test the effects of applying Gaussian filters to the foot x-ray before creating a bone mask.

Instructions

**100 XP**

- Convolve
`im`

with Gaussian filters of size`sigma=1`

and`sigma=3`

. - Plot the "bone masks" of
`im`

,`im_s1`

, and`im_s3`

(i.e., where intensities are greater than or equal to`145`

).