Getting Started

Installation

The purpose of this guide is to show how to get started with using the cuvis SDK. While most of the examples here will be shown in Python, the cuvis SDK is also available in other languages.

These languages include: - C API, usable by many other languages that support C-Style bindings,

but requires more lines of code and offers less convenience.

C++ API, provides the a similar class-oriented design to the Python API, but is usable within C++ and uses the STL where applicable.
Deprecated: C-# and Matlab. While they are available on Github, they are no longer guaranteed to be actively maintained.

Installation of the SDK

Run the latest installer from here.
Ensure the environment variable CUVIS is set to your SDK root.

Installation of the Python API

Recommended: Set up a virtual environment.
Install via pip install cuvis

You can now test the installation via:

import cuvis
print(cuvis.version())

This should now print the version of the installed SDK. You are set up to get started.

Python API: Quick Start

Initialization

import cuvis
cuvis.init("path/to/settings")

[`cuvis.init`](#): Initializes the SDK with the provided user settings directory. [`cuvis.shutdown`](#): Cleanly shuts down the SDK. [`cuvis.set_log_level`](#): Set verbosity of logging, e.g., “info”, “debug”.

Calibration

[`Calibration`](#): Load calibration data from a directory or session file.

calib = cuvis.Calibration("path/to/calibration")

Acquisition

:ref:`AcquisitionContext`_: Manages real-time image acquisition from the camera.

acq = cuvis.AcquisitionContext(calib)
acq.integration_time = 100
mesu = acq.capture_at(500)

[`Component`](#): Access metadata and settings for each hardware component. .. code-block:: python

for comp in acq.components():
print(comp.info.display_name, comp.online)

Processing

[`ProcessingContext`](#): Apply calibration and processing (e.g. Reflectance) to raw measurements.

pc = cuvis.ProcessingContext(mesu)
pc.processing_mode = cuvis.ProcessingMode.Reflectance
pc.apply(mesu)

Measurement and Data

[`Measurement`](#): Encapsulates a single captured or loaded dataset.

session = cuvis.SessionFile("file.cu3s")
mesu = session[0]
cube = mesu.cube

[`ImageData`](#): Handles spectral image arrays and metadata.

spectrum = cube[120, 200][0]  # values at x=120, y=200

[`SessionFile`](#): Load and iterate over measurements in a session file.

session = cuvis.SessionFile("video.cu3s")
for mesu in session:
    print(mesu.name)

Exporting

[`CubeExporter`](#), [`TiffExporter`](#), [`EnviExporter`](#), [`ViewExporter`](#): Export processed measurements to various formats.

from cuvis import TiffExportSettings, TiffExporter
settings = TiffExportSettings(export_dir="output")
exporter = TiffExporter(settings)
exporter.apply(mesu)

Worker

[`Worker`](#): High-level interface to manage acquisition, processing, and exporting in a pipeline.

worker = cuvis.Worker(cuvis.WorkerSettings())
worker.set_acquisition_context(acq)
worker.set_processing_context(pc)
worker.set_exporter(exporter)
worker.start_processing()

Viewer

[`Viewer`](#): Renders view data from measurements using user-defined plugins.

viewer = cuvis.Viewer(settings)
images = viewer.apply(mesu)