# AMMonitor Camera Traps: 1.2M Images from Ten New England Wildlife Monitoring Projects

## Facts

- Provider: AMMonitor community (Clarfeld L., Donovan T. M. et al.); hosted by LILA BC
- License: Public Domain (https://lila.science/datasets/ammonitor-camera-traps/)
- Format: JPEG (in sampled files)
- Stored at: Azure Blob Storage, https://lilawildlife.blob.core.windows.net/lila-wildlife/ammonitor-camera-traps
- Access: public, readable without an account
- Category: Life Sciences
- Source: https://lila.science/datasets/ammonitor-camera-traps/
- Page: https://scigantic.com/archive/dfc26eb5-4171-5dd0-aaad-2fd4d5da555a
- Keywords: Life Sciences, camera traps, JPEG, Azure Blob Storage

## About

A composite of ten camera-trap datasets from the AMMonitor community, independent projects in New England that monitor wildlife with trail cameras and audio recorders using the AMMonitor R package, each first released as a U.S. Geological Survey data release on ScienceBase. Sources include Maine Department of Inland Fisheries and Wildlife (two volumes, one focused on moose), the SiMPL Wildlife Magnet small-mammal project (281,258 media files), Silvio O. Conte National Fish and Wildlife Refuge, Vermont Fish and Wildlife, New Hampshire Fish and Game, Massachusetts wildlife, Dartmouth College Woodlands, and the Green Mountain and White Mountain National Forests. The composite holds about 1.2 million images, roughly half empty; the commonest classes are moose (232,913 images), white-tailed deer (115,240) and fisher (47,505). Annotations are in COCO Camera Traps format with a dataset field per image and location ids; coordinates were removed and images with humans dropped. One folder per source project plus a 45 MB metadata zip. Useful for MegaDetector and species classifier training on northeastern US fauna. The constituent datasets are public domain USGS releases; cite each source data release listed on the LILA page.

## Load the data

```python
import adlfs
fs = adlfs.AzureBlobFileSystem(account_name='lilawildlife', anon=True)
fs.ls('lila-wildlife/ammonitor-camera-traps/')[:10]
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Data structure

File types in sampled files: JPEG

Example file: `dartmouth-woodlands-vol1/Dart 10_Something_20181223_191250_10.JPG`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/dfc26eb5-4171-5dd0-aaad-2fd4d5da555a
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/dfc26eb5-4171-5dd0-aaad-2fd4d5da555a
- Over MCP: `get_schema_card` with id `dfc26eb5-4171-5dd0-aaad-2fd4d5da555a` at https://api.scigantic.com/mcp

## Related topics

- [JPEG datasets](https://scigantic.com/topics/jpeg)
- [Life Sciences datasets](https://scigantic.com/topics/life-sciences)
- [Camera traps datasets](https://scigantic.com/topics/camera-traps)
