This is the hub for the Botany 2023 workshop "Using deep learning with digitized herbarium specimen image data""
The rapid increase in digitized herbarium specimens available in natural history collections has enabled us to study many aspects of plant biology, such as morphological comparisons, shifts in phenology, and changes in species distributions. Currently, tens of millions of herbarium specimen images are available via online databases, with more image data added every day. In recent years, machine learning methods have shown promise in efficiently extracting data from herbarium specimens. Deep learning analyses, such as convolutional neural networks, leverage computer vision to automate analyses of herbarium sheet images. Researchers can use deep learning methods to classify digitized herbarium specimens by species or other user-specified categories. Even relatively simple neural networks can contain hundreds of billions of parameters and require the use of graphical processing units (GPUs), which can process multiple computations simultaneously. Accessing and interfacing with GPUs can represent a barrier to entry for some users. This workshop will help participants to vault past the initial technical challenges of using deep learning on herbarium sheets to conduct analyses for object detection, image segmentation, and taxonomic classification. Participants will work with curated sets of images, or they can use their own datasets--although this will require some advance preparation by the participant to organize their images. The workshop will be conducted in Jupyter notebooks, and some knowledge of Python will be helpful, but not required. A computer with a stable internet connection and a Google account will be required, as well as ~1 GB of free space on Google Drive. Given the hybrid-virtual nature of Botany 2023, conference attendees could also observe the workshop without participating in the hands-on aspects.
There are six sessions, five of which will be hands-on. There is a separate directory in this repository for each section.
- Case studies: how are researchers using deep learning?
- Python coding and using jupyter notebooks on Google Colab
- Acquiring and manipulating digitized herbarium images
- Image data processing and manipulation
- Creating annotated datasets
- Applications of deep learning
Workshop will be held from 8am - 5pm on Sunday, July 23rd, 2023. Workshop attendees must register for and enter this event through the Botany conference platform. The schedule has been modified as of 7/22/2023 because of travel delays.
| Time (in MDT) | Topic |
|---|---|
| 8:00 - 8:15 | Introduction |
| 8:15 - 9:00 | Module 1: Case studies illustrating how researchers are using deep learning (presenter: Rebecca Dikow) |
| 9:10 - 9:55 | Module 2: Introduction to python, jupyter notebooks, and google colab (presenter: Jenna Ekwealor) |
| 10:00 - 10:30 | Coffee break |
| 10:30 - 11:30 | Module 4: Data processing and manipulation options (presenter: Richie Hodel) |
| 12:00 - 1:00 | LUNCH |
| 1:00 - 1:50 | Module 3: Acquiring and manipulating digitized herbarium images (presenter: Sundre Winslow) |
| 2:00 - 2:50 | Module 5: Creating annotated datasets and How to train an object classifier (presenter: Will Weaver) |
| 3:00 - 3:30 | Coffee break |
| 3:30 - 4:00 | Module 6: Example application (presenter: Richie Hodel) |
Etherpad for discussion today:
https://etherpad.wikimedia.org/p/Botany2023_DLworkshop
Botany Code of Conduct:
https://botanyconference.org/file.php?file=SiteAssets/2022_code_of_conduct.pdf
Link to open jupyter notebooks from github in google colab: