Evaluation of edge detectors using average risk

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

10 Citations (Scopus)
23 Downloads (Pure)

Abstract

A new method for evaluation of edge detectors, based on the average risk of a decision, is discussed. The average risk is a performance measure well-known in Bayesian decision theory. Since edge detection can be regarded as a compound decision making process, the performance of an edge detector is context dependent. Therefore, the application of average risk to edge detection is non-trivial. The paper describes a method to estimate the probabilities on a number of different types of (context dependent) errors. A weighted sum of these estimated probabilities represents the average risk. The weight coefficients define the cost function. The method is suitable, not only for the comparison of edge operators, but also for the determination of the weaknesses and strengths of a certain edge operator. This is demonstrated with some well-known edge operators
Original languageEnglish
Title of host publicationProceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages771-774
Number of pages4
ISBN (Print)0-8186-2920-7
DOIs
Publication statusPublished - 30 Aug 1992
Event11th International Conference on Pattern Recognition 1992 - The Hague, the Netherlands, The Hague, Netherlands
Duration: 30 Aug 19923 Sep 1992
Conference number: 11

Conference

Conference11th International Conference on Pattern Recognition 1992
Abbreviated titleICPR 1992
CountryNetherlands
CityThe Hague
Period30/08/923/09/92
Other30 August - 3 September 1992

Fingerprint

Detectors
Edge detection
Decision theory
Cost functions
Decision making

Keywords

  • IR-16506
  • METIS-113391

Cite this

Spreeuwers, L. J., & van der Heijden, F. (1992). Evaluation of edge detectors using average risk. In Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis (pp. 771-774). Piscataway, NJ: IEEE. https://doi.org/10.1109/ICPR.1992.202100
Spreeuwers, Lieuwe Jan ; van der Heijden, Ferdinand. / Evaluation of edge detectors using average risk. Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis. Piscataway, NJ : IEEE, 1992. pp. 771-774
@inproceedings{04aefe61339a455186d4a1ee6a6a798d,
title = "Evaluation of edge detectors using average risk",
abstract = "A new method for evaluation of edge detectors, based on the average risk of a decision, is discussed. The average risk is a performance measure well-known in Bayesian decision theory. Since edge detection can be regarded as a compound decision making process, the performance of an edge detector is context dependent. Therefore, the application of average risk to edge detection is non-trivial. The paper describes a method to estimate the probabilities on a number of different types of (context dependent) errors. A weighted sum of these estimated probabilities represents the average risk. The weight coefficients define the cost function. The method is suitable, not only for the comparison of edge operators, but also for the determination of the weaknesses and strengths of a certain edge operator. This is demonstrated with some well-known edge operators",
keywords = "IR-16506, METIS-113391",
author = "Spreeuwers, {Lieuwe Jan} and {van der Heijden}, Ferdinand",
year = "1992",
month = "8",
day = "30",
doi = "10.1109/ICPR.1992.202100",
language = "English",
isbn = "0-8186-2920-7",
pages = "771--774",
booktitle = "Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis",
publisher = "IEEE",
address = "United States",

}

Spreeuwers, LJ & van der Heijden, F 1992, Evaluation of edge detectors using average risk. in Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis. IEEE, Piscataway, NJ, pp. 771-774, 11th International Conference on Pattern Recognition 1992, The Hague, Netherlands, 30/08/92. https://doi.org/10.1109/ICPR.1992.202100

Evaluation of edge detectors using average risk. / Spreeuwers, Lieuwe Jan; van der Heijden, Ferdinand.

Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis. Piscataway, NJ : IEEE, 1992. p. 771-774.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Evaluation of edge detectors using average risk

AU - Spreeuwers, Lieuwe Jan

AU - van der Heijden, Ferdinand

PY - 1992/8/30

Y1 - 1992/8/30

N2 - A new method for evaluation of edge detectors, based on the average risk of a decision, is discussed. The average risk is a performance measure well-known in Bayesian decision theory. Since edge detection can be regarded as a compound decision making process, the performance of an edge detector is context dependent. Therefore, the application of average risk to edge detection is non-trivial. The paper describes a method to estimate the probabilities on a number of different types of (context dependent) errors. A weighted sum of these estimated probabilities represents the average risk. The weight coefficients define the cost function. The method is suitable, not only for the comparison of edge operators, but also for the determination of the weaknesses and strengths of a certain edge operator. This is demonstrated with some well-known edge operators

AB - A new method for evaluation of edge detectors, based on the average risk of a decision, is discussed. The average risk is a performance measure well-known in Bayesian decision theory. Since edge detection can be regarded as a compound decision making process, the performance of an edge detector is context dependent. Therefore, the application of average risk to edge detection is non-trivial. The paper describes a method to estimate the probabilities on a number of different types of (context dependent) errors. A weighted sum of these estimated probabilities represents the average risk. The weight coefficients define the cost function. The method is suitable, not only for the comparison of edge operators, but also for the determination of the weaknesses and strengths of a certain edge operator. This is demonstrated with some well-known edge operators

KW - IR-16506

KW - METIS-113391

U2 - 10.1109/ICPR.1992.202100

DO - 10.1109/ICPR.1992.202100

M3 - Conference contribution

SN - 0-8186-2920-7

SP - 771

EP - 774

BT - Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis

PB - IEEE

CY - Piscataway, NJ

ER -

Spreeuwers LJ, van der Heijden F. Evaluation of edge detectors using average risk. In Proceedings 11th IAPR International Conference on Pattern Recognition, 1992. Vol.III. Conference C: Image, Speech and Signal Analysis. Piscataway, NJ: IEEE. 1992. p. 771-774 https://doi.org/10.1109/ICPR.1992.202100