Fundamental bounds on edge detection: an information theoretic evaluation of different edge cues
- 20 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 573-579 Vol. 1
- https://doi.org/10.1109/cvpr.1999.786996
Abstract
We treat the problem of edge detection as one of statistical inference. Local edge cues, implemented by filters, provide information about the likely positions of edges which can be used as input to higher-level models. Different edge cues can be evaluated by the statistical effectiveness of their corresponding filters evaluated on a dataset of 100 presegmented images. We use information theoretic measures to determine the effectiveness of a variety of different edge detectors working at multiple scales on black and white and color images. Our results give quantitative measures for the advantages of multi-level processing, for the use of chromaticity in addition to greyscale, and for the relative effectiveness of different detectors.Keywords
This publication has 8 references indexed in Scilit:
- Efficient optimization of a deformable template using dynamic programmingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Elements of Information TheoryPublished by Wiley ,2001
- Minimax Entropy Principle and Its Application to Texture ModelingNeural Computation, 1997
- Perception as Bayesian InferencePublished by Cambridge University Press (CUP) ,1996
- Pattern Recognition and Neural NetworksPublished by Cambridge University Press (CUP) ,1996
- An active testing model for tracking roads in satellite imagesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1996
- Filtering, Segmentation and DepthLecture Notes in Computer Science, 1993
- A Computational Approach to Edge DetectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1986