Edge Detector Evaluation Using Empirical ROC Curves

Abstract

A method is demonstrated to evaluate edge detector performance using receiver operating characteristic curves. It involves matching edges to manually specified ground truth to count true positive and false positive detections. Edge detector parameter settings are trained and tested on different images, and aggregate test ROC curves presented for two sets of 10 images. The performance of eight different edge detectors is compared. The Canny and Heitger detectors provide the best performance.

Cite

Text

Bowyer et al. "Edge Detector Evaluation Using Empirical ROC Curves." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1999. doi:10.1109/CVPR.1999.786963

Markdown

[Bowyer et al. "Edge Detector Evaluation Using Empirical ROC Curves." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1999.](https://mlanthology.org/cvpr/1999/bowyer1999cvpr-edge/) doi:10.1109/CVPR.1999.786963

BibTeX

@inproceedings{bowyer1999cvpr-edge,
  title     = {{Edge Detector Evaluation Using Empirical ROC Curves}},
  author    = {Bowyer, Kevin W. and Kranenburg, Christine and Dougherty, Sean},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {1999},
  pages     = {1354-1359},
  doi       = {10.1109/CVPR.1999.786963},
  url       = {https://mlanthology.org/cvpr/1999/bowyer1999cvpr-edge/}
}