Published March 3, 2021 | Version v1
Dataset Open

RCH Detection with Mask R-CNN Image Annotations

  • 1. Carleton University
  • 2. Muhlenberg College
  • 3. Lehigh University

Description

These data are part of a project to detect relict charcoal hearths (RCH) in landscape images using Mask R-CNN, a deep learning library based on Keras and Tensorflow.

This is a collection of xml files that contain coordinates of the locations of known RCHs on images. These files are known as annotations and are used by Mask R-CNN to identify objects to detect during training of a model.

For additional information, please see https://zenodo.org/record/4766351

Files

annots-20210303T113626Z-001.zip

Files (468.2 kB)

Name Size Download all
md5:0786ddae3d81230916b06020d2ff12c4
468.2 kB Preview Download

Additional details

Related works

Is documented by
10.5281/zenodo.4766351 (DOI)