Dataset of images of wildlife taken by motion sensor cameras on UK farms

The image dataset is for image recognition purposes. The images were provided by our funder: The Department for Environment, Farming and Rural Affairs.

This newly created image dataset contains 8368 images, which we manually resized, categorized and labelled. 
The images belong to one of six categories: badger, bird, cat, fox, rat and rabbit.

Seventy percent of the images in each class were randomly chosen as the training dataset, the rest were assigned to the testing dataset.

The training folder contains training images and the testing folder contains testing images.

This software is linked with the publication: 
Chen, R, Little, R, Mihaylova, L, Delahay, R, Cox, R. Wildlife surveillance using deep learning methods,
 Ecology and Evolution, 2019, Vol. 9, pp. 9453– 9466. https://doi.org/10.1002/ece3.5410