Principal Component Analysis (PCA)¶

PCA is a technique used to emphasize variation and bring out strong patterns in a dataset. It's often used to make data easy to explore and visualize. In this notebook, we'll perform PCA on a multi-spectral imagery to reduce the dimensionality of the data and visualize the results.

In [1]:

import hypercoast

import hypercoast

In [2]:

url = "https://github.com/opengeos/datasets/releases/download/hypercoast/Planet_2022-09-23.tif"
filepath = "data/Planet_2022-09-23.tif"
hypercoast.download_file(url, filepath)

url = "https://github.com/opengeos/datasets/releases/download/hypercoast/Planet_2022-09-23.tif" filepath = "data/Planet_2022-09-23.tif" hypercoast.download_file(url, filepath)

Out[2]:

'/home/runner/work/HyperCoast/HyperCoast/docs/examples/data/Planet_2022-09-23.tif'

In [3]:

output_file = filepath.replace(".tif", "_pca.tif")

output_file = filepath.replace(".tif", "_pca.tif")

In [4]:

hypercoast.pca(filepath, output_file, n_components=3)

hypercoast.pca(filepath, output_file, n_components=3)

In [5]:

m = hypercoast.Map()
m.add_basemap("Hybrid")
m.add_dataset(
    filepath, indexes=[6, 4, 2], vmin=0, vmax=2500, layer_name="Planet Image", nodata=0
)
m.add_dataset(
    output_file, indexes=[0, 1, 2], vmin=0, vmax=2500, layer_name="PCA Image", nodata=0
)
m.add("spectral")
m

m = hypercoast.Map() m.add_basemap("Hybrid") m.add_dataset( filepath, indexes=[6, 4, 2], vmin=0, vmax=2500, layer_name="Planet Image", nodata=0 ) m.add_dataset( output_file, indexes=[0, 1, 2], vmin=0, vmax=2500, layer_name="PCA Image", nodata=0 ) m.add("spectral") m

Out[5]: