Guide

Contents:

• Image Operations
  • Indexing
  • Elementwise operations
  • Reduction operations
  • Resolving calculations
• Preprocessing
  • Reflectance transformation
• Machine Learning
  • PCA
• Custom operations
  • General operations
  • Machine learning models
• Recipes
  • Spectral calibration
  • Scatter correction (SNV)
  • Mean spectrum
  • PCA from random sample

Reading and writing images

Basic file inspection and format conversion is done using the open and write functions of the library.

PythonRust

import hsi

img = hsi.open("path/to/file.hdr") # Expects ENVI file due to the extension
img.write("path/to/output.pam") # Writes PAM format

The file object can be inspected to get information about the image size and metadata.

print(img.shape)
print(img.header.interleave)
print(img.header.wavelengths)

Interacting with NumPy

The hsi.HSImage provides conversions to/from NumPy. The memory layout of the file or image determines the layout of the ndarray that is returned or used as input (see Memory layout and interleave for details on the layout of the ndarray that is returned or used as input (see Memory layout and interleave for details on the memory layout of HSI files).

Conversions to NumPy can be done directly using the hsi.HSImage.to_numpy() method.

a = hsi.open("path/to/file.hdr").to_numpy()

The interleave can be changed using hsi.HSImage.to_numpy_with_interleave():

a = hsi.open("path/to/file.hdr").to_numpy_with_interleave(hsi.bil)

You can also create a new hsi.HSImage object directly from a NumPy array:

img = hsi.HSImage.from_numpy(a, interleave=hsi.bsq)