Title :
On the localization performance measure and optimal edge detection
Author :
Tagare, Hemant D. ; DeFigueiredo, Rui J P
Author_Institution :
Dept. of Electr. Eng., Rice Univ., Houston, TX, USA
fDate :
12/1/1990 12:00:00 AM
Abstract :
The localization performance measure of edge detectors is addressed. A one-dimensional formulation of the problem is considered. A linear space-invariant filter is used for the detection. The locations of local maxima in the thresholded output of the filter are declared to be the edges. The limitations of conventional performance measures are shown, and a localization performance measure for edge detection is suggested. This performance measure is based on the theory of zero-crossings of stochastic processes. It is shown that the derivative of a Gaussian is the optimal edge detector for the measure
Keywords :
filtering and prediction theory; optimisation; pattern recognition; picture processing; stochastic processes; edge detection; linear space-invariant filter; local maxima; localization performance measure; pattern recognition; stochastic processes; zero-crossings; Biomedical imaging; Computed tomography; Degradation; Detectors; Image edge detection; Image processing; Medical diagnostic imaging; Nonlinear filters; Radiology; Stochastic processes;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
