Hi everyone!
It is time for another instalment of the monthly paper round-ups for readers of the open source solutions feed.
This is not necessarily an endorsement of any of the presented work and this list is not intended to be comprehensive. If you think I missed something good, please share in the comments!
Here are the picks for August 2024.
August 2024 paper round-up
HyperCoast: A Python Package for Visualizing and Analyzing Hyperspectral Data in Coastal Environments
HyperCoast is a Python package for working with various types of hyperspectral data on coastal areas. Features include searching hyperspectral data by NASA, performing atmospheric corrections and various interactive visualisation tools for different types of data.
Paper: https://doi.org/10.21105/joss.07025
Project: https://hypercoast.org/
Fluxbot: The Next Generation - Design and Validation of a Wireless, Open-Source Mechatronic CO2 Flux Sensing Chamber
This paper presents a generational update to Fluxbot, a low-cost CO2 flux sensor for soil. Fluxbot 2.0 is capable of real time wireless data transmission, costs under USD 500 and has a ten times longer battery life (over 300 hours).
Paper: https://doi.org/10.1145/3674829.3675063
Project: https://github.com/elizabethforbes/fluxbots2019_archive/tree/v1.0.0
BirdVoxDetect: Large-Scale Detection and Classification of Flight Calls for Bird Migration Monitoring
This open source code contains three machine learning modules: a random forest classifier for sensor faults, a deep convolutional neural network for sound event detection, and finally, a multitask convolutional neural network for predicting family, genus and species. It has been evaluated on 296 hours of recordings from nine locations.
Paper: https://doi.org/10.1109/TASLP.2024.3444486
Project: https://github.com/BirdVox/birdvoxdetect
Publication-ready single nucleotide polymorphism visualization with snipit
Snipit is a Python package for analysing and visualising single nucleotide polymorphisms (SNPs) compared to a reference sequence. This project was originally motivated for comparing SARS-CoV-2 variants, but has since been used for lots of other virus types, such as HIV and the Poliovirus.
Paper: https://doi.org/10.1093/bioinformatics/btae510
Project: https://github.com/aineniamh/snipit
Coastal generalized ecosystem model (CGEM) 1.0: Flexible model formulations for simulating complex biogeochemical processes in aquatic ecosystems
CGEM is a software tool for building water quality models. Users can use it for studying regulating processes of water column optical properties, phytoplankton and zooplankton dynamics and benthic carbon, oxygen and nutrient cycles.
Paper: https://doi.org/10.1016/j.ecolmodel.2024.110831
Project: https://github.com/USEPA/CGEM
ZebraPose: Zebra Detection and Pose Estimation using only Synthetic Data
This is a working paper on arxiv. The authors used synthetic data generated with a 3D simulator to train pose estimation on 2D images of zebras. They also show that it is possible to generalise the same models to 2D pose estimation of horses with only a small sample of real world images to retrain.
Paper: https://arxiv.org/pdf/2408.10831
Project: https://zebrapose.is.tue.mpg.de/
“The Devil You Know": Barriers and Opportunities for Co-Designing Microclimate Sensors
For this paper, researchers conducted 11 interviews with experts who use environmental sensing tech about the state of microclimate monitoring. They also spent three months observing wild rice conservation work. The paper includes a section on trade-offs between proprietary and open source solutions.
Paper: https://doi.org/10.1145/3685695
Discovering and measuring giant trees through the integration of multi-platform lidar data
This paper presents open source code for detecting and measuring giant trees using lidar and UAVs.
Paper: https://doi.org/10.1111/2041-210X.14401
Project: https://github.com/YuRen1992/Lidar-Segmentation-Fusion
GeoPick - A web application for georeferencing natural history collections following best practices
GeoPick is a web app for assigning location tags to specimens of natural history collections. It has a nice GUI for a user-friendly experience.
Paper: https://doi.org/10.1111/ecog.07431
Project: https://github.com/rtdeb/GeoPick
Add the first post in this thread.