Overview   Tree   Index 
NPR Literature
PREV  NEXT FRAMES  NO FRAME 

[Can86]  A Computational Approach to Edge Detection

Canny:1986:ACA (Article)
Author(s)Canny
Title« A Computational Approach to Edge Detection »
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume8
Number6
Page(s)679--698
Year1986

Abstract
This paper describes a computational approach to edge detection. The success of the approach depends on the definition of a comprehensive set of goals for the computation of edge points. These goals must be precise enough to delimit the desired behavior of the detector while making minimum assumptions about the form of the solution. We define detection and localisation criteria for a class of edges, and present mathematical forms for these criteria as functionals on the operator impulse response. A third criteria is then added to ensure that the detector has only one response to a single edge. We use the criteria in numerical optimisation to derive detectors for several common image features, including step edges. On specialising the analysis to step edges we find that there is a natural uncertainty principle between detection and localisation performance, which are the two main goals. With this principle we derive a single operator shape which is optimal at any scale. The optimal detector has a single approximate implementation in which edges are marked at maxima in gradient magnitude in a Gaussian smoothed image. We extend this simple detector using operators of several widths to cope with different signal to noise ratios in the image. We present a general method called feature synthesis for the fine to coarse integration of information from operators at different scales. Finally we show that step edge detector performance improves considerably as the operator point spread function is extended along the edge. This detection scheme uses several elongated operators at each point, and the directional operator outputs are integrated with the gradient maximum detector.

BibTeX code
@article{Canny:1986:ACA,
  number = {6},
  volume = {8},
  optnote = {CANNY86},
  author = {F. John Canny},
  optstatus = {OK, URL, Paper, RI},
  localfile = {papers/Canny.1986.ACA.pdf},
  title = {A {C}omputational {A}pproach to {E}dge {D}etection},
  abstract = {This paper describes a computational approach to edge detection.
              The success of the approach depends on the definition of a
              comprehensive set of goals for the computation of edge points.
              These goals must be precise enough to delimit the desired behavior
              of the detector while making minimum assumptions about the form of
              the solution. We define detection and localisation criteria for a
              class of edges, and present mathematical forms for these criteria
              as functionals on the operator impulse response. A third criteria
              is then added to ensure that the detector has only one response to
              a single edge. We use the criteria in numerical optimisation to
              derive detectors for several common image features, including step
              edges. On specialising the analysis to step edges we find that
              there is a natural uncertainty principle between detection and
              localisation performance, which are the two main goals. With this
              principle we derive a single operator shape which is optimal at
              any scale. The optimal detector has a single approximate
              implementation in which edges are marked at maxima in gradient
              magnitude in a Gaussian smoothed image. We extend this simple
              detector using operators of several widths to cope with different
              signal to noise ratios in the image. We present a general method
              called feature synthesis for the fine to coarse integration of
              information from operators at different scales. Finally we show
              that step edge detector performance improves considerably as the
              operator point spread function is extended along the edge. This
              detection scheme uses several elongated operators at each point,
              and the directional operator outputs are integrated with the
              gradient maximum detector.},
  doi = {http://doi.acm.org/10.1145/11274.11275},
  journal = j-IEEE-PAMI,
  pages = {679--698},
  year = {1986},
}

 Overview   Tree   Index 
NPR Literature
PREV  NEXT FRAMES  NO FRAME 

Submit a bug

This document was generated by bib2html 3.3.
Copyright © 1998-05 Stéphane GALLAND (under the GNU General Public License)

Valid HTML 4.01!Valid CSS!