Theory of edge detection

By D. Marr, Ellen C. Hildreth. 1979

Subfields: Computer Vision and Pattern Recognition, Algorithms & Complexity

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Algorithms & Complexity

Citations in the corpus

Holistically-Nested Edge Detection (2015, methodology): “It has been long understood that precisely localizing edges in natural images involves visual perception of various “levels” [16, 25].”
Machines that learn to segment images: a crucial technology for connectomics. (2010, background): “Local boundary detectors generally look for abrupt changes in various properties such as intensity, color, and texture [20,21,8].”
Nonlocal Metasurfaces for Optical Signal Processing. (2018, background): “Our results pave the way towards the realization of efficient analog signal processing devices for edge detection and image processing operations [31],[32].”
XDoG: An eXtended difference-of-Gaussians compendium including advanced image stylization (2012, unknown): “Marr and Hildreth suggested circumventing this problem by using the Laplacian ∇ 2 = ∂ 2 ∂x 2 + ∂ 2 ∂y 2 .”
Medical image registration: a review (2014, methodology): “Usual border-based segmentation techniques include edge detectors based on image gradient (Marr and Hildreth 1980; Canny 1986), corner detectors, line detectors based on the Hough transform; deformable models, like active contours, usuallyknownassnakes(Kassetal.1988; CootesandTaylor 1992; McInerney…”
Objective comparison of particle tracking methods (2014, methodology): “METHOD 4: PARAMETERS The user parameters of this method (and typical values) are:  Number of iterations of convolution with the Gaussian kernel (10-500).”
Fractional differentiation for edge detection (2003, methodology): “In image processing, edge detection often makes use ofinteger-order di�erentiation operators, especially order 1 used by the gradient and order 2 by the Laplacian [2,4,5,10,13, 16 ,25,26,28,31].”
Plasmonic computing of spatial differentiation (2017, methodology): “Equation (2) is the transfer function of a first-order spatial differentiator.”
Relation learning in a neurocomputational architecture supports cross-domain transfer (2019, methodology): “it because it is adequate for the current purposes, it is computationally inexpensive, and its operations are consistent with what is known about early vision (e.g., detecting edges by local contrast [55] and that absolute extent information in the world is represented homomorphically in the visual ”
Flux maximizing geometric flows (2001, background): “However, for the special case that V is the gradient of a potential function such as an intensity image I, the flux maximizing flow has a very interesting connection to Marr and Hildreth’s work on edge detection [25].”
Exploring photobios (2011, background): “Real image edge profiles tend to be smooth rather than discontinuous, due to the optical blurring effects of the imaging process [Marr and Hildreth 1980; Nalwa and Binford 1986].”
Magnetic resonance image tissue classification using a partial volume model. (2001, methodology): “We locate the anatomical boundaries in MRI brain volumes using the Marr–Hildreth edge detector (Marr and Hildreth, 1980).”