Launch binder

Smooth an image

This example shows how to use to smooth an image layer of

We use the argument method="smooth" to smooth the image. This calls skimage.filters.gaussian() in the background. Keyword arguments kwargs are passed to the wrapped function. This allows us to set the width of the Gaussian kernel, \(\\sigma\), used for smoothing.

import squidpy as sq

import matplotlib.pyplot as plt

# load the H&E stained tissue image
img = sq.datasets.visium_hne_image_crop()

Smooth the image with sigma = 2. With the argument layer we can select the image layer that should be processed. By default, the resulting image is saved in the layer image_smooth. This behavior can be changed with the arguments copy and layer_added., layer="image", method="smooth", sigma=2)

Now we can look at the result on a cropped part of the image.

crop = img.crop_corner(0, 0, size=200)

fig, axes = plt.subplots(1, 2)
for i, layer in enumerate(["image", "image_smooth"]):, ax=axes[i])
image, image_smooth

Total running time of the script: ( 0 minutes 9.585 seconds)

Estimated memory usage: 43 MB