计算机视觉CV6.pptVIP

Computational Vision CSCI 363, Spring 2008Lecture 6Edge Detection Why Find Edges? Human Psychophysics Stages for Edge Detection Smoothing Localization of an edge Intensity Derivative Zero Crossings and Edges Convolution The Derivative of a Convolution A 2D Gaussian Difference of Gaussians * Want to find meaningful objects or surfaces Usually the border of an object is defined by a change in light intensity. Changes in intensity can also signal a change in depth or orientation of the surface. Must first find the intensity changes. Then must find what led to the intensity change. Humans can detect sharp intensity changes: Craik-Obrien-Cornsweet illusion #1 Intensity Distance Demos: (Viperlib.org) Humans are not good at detecting gradual intensity changes. Detecting Edges: Smoothing--Eliminates noise. Determines spatial scale. Differentiation--Localizes the intensity change Feature Extraction: Determine the feature that caused the intensity change. Intensity Smoothing More Smoothing Changes in intensity are not instantaneous, particularly in a smoothed image. Humans can localize edges to within a few seconds of arc (a few mm for a line a distance of 1 meter from the observer). Accurate localization is necessary for stereo vision. Differentiation allows us to find the location of the most rapid intensity change. The first derivative gives a peak at the location of the most rapid change. The second derivative gives a zero at this location. Marr and Hildreth suggested using these zero crossings to indicate edges. Smoothed Intensity First Derivative Second Derivative Image Image after smoothing and second derivative Black = Negative White = Positive Zero Crossings To smooth a function, we replace the value at each x position with the averaged values around that position. This process is called convolution. Smoothing works best if the average is weighted: nearby neighbors count more, distant neighbors count less. A Gaussian function works well for this weightin