Sorting the backlog of physical butterfly and moth specimens at the ETH Entomological Collection will require the assistance of staff and volunteers who have no prior training in species identification. To enable these individuals to contribute effectively, they require a technological aid to allow them to differentiate between the different hierarchical groupings of butterflies and moths. This will allow the workload for subject experts to be significantly reduced.
A publicly available mobile application for non-expert entomological collection workers, which suggests the most likely species for each specimen of butterfly or moth photographed. This will allow the non-expert staff to correctly sort the majority of specimens which can later be quickly checked by an expert.
- Built for iOS and Android
- Tech Stack
- React-Native with Expo framework
- This API is is used for user management as well as storing data about prediction cases. This data could be used to retrieve all previous cases, analysis etc.
- Tech Stack
- Django and Django REST Framework
- Postgres
- Redis
- Celery
- Teck Stack
- Docker
- Docker Compose
- Nginx
- Gunicorn
- Gitlab CI/CD
The predictions API is in a separate codebase. It is used in this project to send a picture of a butterfly and get the related predictions in return.