This article is within the scope of WikiProject Robotics, a collaborative effort to improve the coverage of Robotics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.RoboticsWikipedia:WikiProject RoboticsTemplate:WikiProject RoboticsRobotics
This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing
This article is within the scope of WikiProject Computer science, a collaborative effort to improve the coverage of Computer science related articles on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Computer scienceWikipedia:WikiProject Computer scienceTemplate:WikiProject Computer scienceComputer science
In the practice of digital image enhancement, basing edge detection merely
on numerical derivatives is too naive, and unrealistic. For each pixel of
a digital image, one wants not only to decide if it is a candidate for membership in an "edge" but also to find the direction of that edge.
[In particular, the edge direction is required for true sharpening].
One needs to analize a suitable collection of neighboring pixels (typically, those at horizontal and vertical distances up to 3) with respect
to intensity as well as position. Although effective methods of doing this
are not very difficult to develop, it seems that commercial software
does not provide truly suitable implementations.
Response to unsigned criticism above: Well, edge detection based on image derivatives is not fully naive, subject to the well-known practice of using Gaussian filtering as a pre-processing stage to the computation of image derivatives. This means that the effective support region for image derivative computations are equal to the support regions of first-order Gaussian derivative operators, and thus substantially larger than a distance of three pixels. Moreover, the orientation of an edge within the differential approach to edge detection is given as orthogonal to the orientation of the image gradient as estimated by first-order Gaussian derivative operators. In practice, these approaches have found numerous successful applications in computer vision, however, usually with different goals than mere image enhancement. Tpl
Following the tag marked in October 2007, I have now made a first attempt to restructure this article to be more updated with respect to the topic of edge detection and also to give more technical details of basic edge detectors. Question to those of you who have tagged this article, do you find it appropriate to remove the tag? Tpl (talk) 16:40, 22 February 2008 (UTC)[reply]
.jpg?lang=en){kind=small}
