A curated list of open technology projects to sustain a stable climate, energy supply, biodiversity and natural resources.

Recent Releases of EcoAssist

EcoAssist - AddaxAI v6.10

  • Fix bug for incomplete AddaxAI JSON conversion
  • Update SpeciesNet params properly at initiation

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 23 days ago

EcoAssist - AddaxAI v6.09

  • Google's global species identification model added

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 26 days ago

EcoAssist - AddaxAI v6.08

  • Solved GPU not found on Windows NVIDIA GPUs
  • Added Sensing Clues export format TSV

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 28 days ago

EcoAssist - AddaxAI v6.07

  • Enables downloading of virtual environments during runtime to reduce size
  • Fix 'ModuleNotFoundError: No module named 'tensorflow_addons'' error when running Peruvian Amazon model

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 29 days ago

EcoAssist - AddaxAI v6.06

  • Allow full image classifiers

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 month ago

EcoAssist - AddaxAI v6.05

  • Fixed bug leading to "UnboundLocalError"

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 month ago

EcoAssist - AddaxAI v6.04

  • Fixed bug leading to "No such file or directory"
  • Fixed bug leading to "UnidentifiedImageError"

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 month ago

EcoAssist - AddaxAI v6.03

  • New European model (v1.3)

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 2 months ago

EcoAssist - AddaxAI v6.02

  • Allow installation for all users as opposed to single user only
  • Fix issue with COCO format and images without metadata
  • Added functionality to blur people from images
  • Made AddaxAI backwards compatible with JSON files created with EcoAssist versions

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 2 months ago

EcoAssist - AddaxAI v6.01

To avoid any legal concerns, we have renamed the project from EcoAssist to AddaxAI. The project itself remains the same—only the name has changed. The new website is: addaxdatascience.com/addaxai

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 2 months ago

EcoAssist - EcoAssist v5.29

  • New input data formats:
    • images: '.jpg', '.jpeg', '.gif', '.png', '.tif', '.tiff', and '.bmp'
    • videos: '.mp4', '.avi', '.mpeg', '.mpg', '.mov', and '.mkv'
  • New feature to select line width and font size for image visualization (XS, S, M, L, XL)

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.28

  • Added CV4Ecology workshop inference code
  • Removed redundant EXE files to avoid anti virus reports

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.27

Bug fix for Windows users and the human-in-the-loop annotation feature.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.26

This version has:

  • adjusted environment names for Linux users
  • added local install of PyQt5 and pyrcc5 command for human in the loop feature
  • added open.command for Linux users

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.25

Truncated images can now be read and analysed by the classification models.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.24

And again fixed some additional bugs ;)

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.23

Fixed some initial bugs related to the new automated install.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.22

This release incorporates a new method of installation that hopefully bypasses restrictions on locked down machines. It removes the dependency for conda and uses a graphical user interface during the installation.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 3 months ago

EcoAssist - EcoAssist v5.21

Enhanced video handling capabilities and integration with the latest video features in Timelapse.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.20...v5.21

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 5 months ago

EcoAssist - EcoAssist v5.20

Beta-version with per-frame video visualisations using Timelapse integration.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.19...v5.20

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 5 months ago

EcoAssist - EcoAssist v5.19

Added the new version of the European species identification model (Deepfaune v1.2).

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.18...v5.19

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 6 months ago

EcoAssist - EcoAssist v5.18

Added new species identification model for Kirghizistan fauna.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.17...v5.18

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 7 months ago

EcoAssist - EcoAssist v5.17

  • Fixed installation bug
  • Fixed problems with Southwest USA model for CPU users

What's Changed

New Contributors

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.16...v5.17

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 7 months ago

EcoAssist - EcoAssist v5.16

Includes a new species identification model for Southwest USA with 27 classes. Developed by the San Diego Zoo Wildlife Alliance.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.15...v5.16

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 8 months ago

EcoAssist - EcoAssist v5.15

Added a check for corrupted files and the option to repair them before starting annotation session.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 8 months ago

EcoAssist - EcoAssist v5.14

New export in COCO format.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 9 months ago

EcoAssist - EcoAssist v5.13

Facilitates the deployment of the new Peruvian Andes model.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 9 months ago

EcoAssist - EcoAssist v5.12

This version fixes the bug that linked incorrect labels to detections in custom MEWC models.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 10 months ago

EcoAssist - EcoAssist v5.11

This update includes the beta version for the Timelapse incorporation. Soon users will be able to deploy models via EcoAssist directly from Timelapse.

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 10 months ago

EcoAssist - EcoAssist v5.10

Supports the deployment of the Peruvian Amazon model.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.4...v5.10

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 10 months ago

EcoAssist - EcoAssist v5.9

  • Fixed bug with NaN datetimes
  • Fixed draw square bbox bug

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.8...v.5.9

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 10 months ago

EcoAssist - EcoAssist v5.8

Added smoothing feature to reduce noise by averaging predictions over images, videos and sequences.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.7...v.5.8

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 10 months ago

EcoAssist - EcoAssist v5.7

Generate plots, graphs and maps based on the detections.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.6...v.5.7

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 11 months ago

EcoAssist - EcoAssist v5.6

This version is compatible with the PytorchWildlife Amazon Rainforest model.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.5...v.5.6

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 11 months ago

EcoAssist - EcoAssist v5.5

Adjusted code to import MEWC class_list.yaml

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.4...v.5.5

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 11 months ago

EcoAssist - EcoAssist v5.4

Fixed conda init bug and finalised NZ model.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.3...v5.4

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren 11 months ago

EcoAssist - EcoAssist v5.3

Adjusted code for the new NZ species identification model.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.2...v5.3

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 year ago

EcoAssist - EcoAssist v5.2

Added DeepFaune v1.1 model to the list of classifiers.

What's Changed

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v5.0...v5.2

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 year ago

EcoAssist - EcoAssist v5.1

Moved PyQt5 dependency to PySide6 for Unix systems to avoid HomeBrew dependency and extra install.

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v.5.0...v5.1

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 year ago

EcoAssist - EcoAssist v5.0

  • new window: simple mode
  • saves user settings
  • incorporated Namibian species model
  • automatic model download
  • switch between virtual environments

What's Changed

New Contributors

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v4.4...v.5.0

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 1 year ago

EcoAssist - EcoAssist v4.4

In previous versions of EcoAssist (v3.0 > v4.3) it was possible to train your own object detection models based on MegaDetector to detect your target species. Although this did work, it wasn't the best approach to develop a species recognition model. It required lots of training data, processing power, time, electricity and wasn't very accurate. Advancing insights revealed that better results can be obtained by using a object classification model to be used in conjunction with the results of MegaDetector. The animals will then be located by MegaDetector, and further classified by your custom model. EcoAssist > v4.2 does support the deployment of a classification model to be used in conjunction with MegaDetector, but training such a model is more complicated and hasn't been incorporated into EcoAssist (yet).

Further:

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v4.3...v4.4

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 1 year ago

EcoAssist - EcoAssist v4.3

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v4.2...v4.3

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 1 year ago

EcoAssist - EcoAssist v4.2

  • Annotation in xml to be able to store more information
  • human in the loop feature
  • minor fixes

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v4.1...v4.2

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 1 year ago

EcoAssist - EcoAssist v4.1

Added human-in-the-loop functionality to verify detections and create training data. Possibility to select an image subset based on specific classes, confidence ranges, and selection methods.

What's Changed

New Contributors

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v4.0...v4.1

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 1 year ago

EcoAssist - EcoAssist v4.0

Added functionality to train object detection models and fixed bugs.

What's Changed

Full Changelog: https://github.com/PetervanLunteren/EcoAssist/compare/v3.0...v4.0

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren almost 2 years ago

EcoAssist - EcoAssist v3.0

  • New interface
  • Postprocessing features work with custom classes
  • Accommodates both new and old yolov5 models
  • Separate confidence thresholds per feature
  • New feature to sort data based on confidence

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren about 2 years ago

EcoAssist - EcoAssist v2.2

  • All features work with relative path
  • New option to copy or move files when separated
  • Fixed bug regarding faulty checkpoint frequency and sample rate
  • Lowered the default threshold for images to 0.1 and videos to 0.2
  • Improved information text about threshold values and further processing in Timelapse
  • Added message box after processing if created .xml files

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 2 years ago

EcoAssist - EcoAssist v2.1

Bug fixes and thoroughly tested.

Features:

  • Use either MDv5a or MDv5b to tag animals, persons and vehicles in both images and video's
  • Filter out empty images, people, vehicles or animals
  • Review and edit annotations using the open-source annotation software labelImg
  • Create input file for further processing in Timelapse
  • Export .xml label files in Pascal VOC format for further model training
  • Manipulate data by drawing boxes or cropping detections
  • Use custom models trained from MDv5 using transfer learning

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 2 years ago

EcoAssist - v2.0

  • adopted to windows, linux and osx
  • suitable for gpu accelerated computers
  • using latest MDv5 model

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 2 years ago

EcoAssist - v1.0

  • suitable for OSx computers
  • using MDv4 model

Biosphere - Terrestrial Wildlife - Python
Published by PetervanLunteren over 2 years ago