Roberts Cross Edge Detector - 2×2 gradient edge detector
Sobel Edge Detector - 3×3 gradient edge detector
Canny Edge Detector - non-maximal suppression of local gradient magnitude
Compass Edge Detector - 3×3 gradient edge detectors
Zero Crossing Detector - edge detector using the Laplacian of
Gaussian operator
Line Detector - line (as distinct from edge) feature detector
The operators included in this section are those whose purpose is to identify meaningful image features on the basis of distributions of pixel graylevels. The two categories of operators included here are:
Edge
Pixel Detectors, that assign a value to a pixel in proportion to
the likelihood that the pixel is part of an image edge
(i.e. a pixel that is on the boundary between two regions of different
intensity values).
Line Pixel
Detectors, that assign a value to a pixel in proportion to the
likelihood that the pixel is part of a image line (i.e. a dark narrow
region bounded on both sides by lighter regions, or vice-versa).
Detectors for other features can be defined, such as circular arc detectors in intensity images (or even more general detectors, as in the generalized Hough transform), or planar point detectors in range images, etc.
Note that the operators in this section merely identify pixels likely to be part of such a structure. To actually extract the structure from the image it is then necessary to group together image pixels (which are usually adjacent). A verification stage also commonly occurs after the grouping stage (e.g. an examination of the residuals of the best fit line through a point set) and then the extraction of parameters of the structure (e.g. the equation of the line through the point set).
©2003 R. Fisher, S. Perkins,
A. Walker and E. Wolfart.
