oplab_pipeline

oplab_pipeline is a python toolchain to process AUV dives from raw data into navigation and imaging products. The software is capable of:

• Process navigation: fuses AUV or ROV sensor data using state of the art filters and geolocalises recorded imagery.
• Camera and laser calibration: performs automatic calibration pattern detection to calibrate monocular or stereo cameras. Also calibrates laser sheets with respect to the cameras.
• Image correction: performs pixel-wise image corrections to enhance colour and contrast in underwater images.

Please review the latest changes in the CHANGELOG.md.

Installation

For production, to install this package run:

pip install -U git+https://github.com/ocean-perception/oplab_pipeline.git

This will make the commands auv_nav, auv_cal and correct_images available in the terminal. For more details refer to the documentation.

For development, clone the repository, navigate to the oplab-pipeline folder and run

pip install -U --user -e .

Notes:

To import rosbag, using pip install baypy. (see the docs: https://jmscslgroup.github.io/bagpy/)

Documentation

The documentation is hosted in read the docs.

Citation

If you use this software, please cite the following article:

Yamada, T, Prügel‐Bennett, A, Thornton, B. Learning features from georeferenced seafloor imagery with location guided autoencoders. J Field Robotics. 2020; 1– 16. https://doi.org/10.1002/rob.21961

License

Copyright (c) 2020-2022, University of Southampton. All rights reserved. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

See LICENSE.md file in the project root for full license information.

Contributing

Please document the code using Numpy Docstrings. If you are using VSCode, there is a useful extension that helps named Python Docstring Generator. Once installed, make sure you select Numpy documentation in the settings.

Run pre-commit install to install pre-commit into your git hooks. pre-commit will now run on every commit. If you don't have pre-commit installed, run pip install pre-commit.