AWhat changed
An edge-detector design balances detection, localization and duplicate responses, leading to an efficient Gaussian-gradient approximation with hysteresis and multiscale integration.
Scientific scope
An edge-detector design balances detection, localization and duplicate responses, leading to an efficient Gaussian-gradient approximation with hysteresis and multiscale integration. The filter criteria use a specified edge cross-section, additive white Gaussian noise and a small-noise localization approximation. Expected peak spacing is not a guarantee of exactly one response. Numerical filter design, Gaussian-derivative approximation, nonmaximum suppression and hysteresis require parameter choices. Scale trades signal-to-noise against localization. The method does not promise universal optimum detection across textures, corners or all noise models; combined edge/ridge results were inconclusive.
Evidence
Source-linked, scoped account of the original work. The published scope states the scientific assumptions and remaining verification limits. WRS has not independently certified the full proof, experimental result or system benchmark.
Contributors
- John Canny · SCIENCE: Complete original scientificbyline;1983report/journal onefamily,noeffortfraction inferred.
- Cited filter/edge predecessors · SCIENCE: Antecedent methods and formulas, distinct from this paper's byline authors.
- J.M.Brady; anonymous referee · SCIENCE: Acknowledged influence, draft feedback and Schwarz derivation; no additional author identity or ownership fraction is inferred.
- SystemDevelopmentFoundation/ONR/ARPA · INFRASTRUCTURE: Source-listed researchfunding;notbylineauthors.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction