Title :
A local orientation coherency weighted color gradient for edge detection
Author_Institution :
Hewlett-Packard Lab., Palo Alto, CA, USA
Abstract :
We present a novel color gradient operator based on a weighted max distance (a weighted Minkowski distance of L∞ norm) with the weights determined by the coherency of local gradient orientations. We demonstrate its performance advantages by quantitatively comparing it with several existing gradient operators using synthetic images.
Keywords :
edge detection; gradient methods; image colour analysis; mathematical operators; matrix algebra; color gradient operator; edge detection; synthetic images; weighted color gradient; weighted max distance; Color; Colored noise; Detectors; Gray-scale; Image edge detection; Laboratories; Milling machines; Retirement; Signal to noise ratio; Smoothing methods; color edge detector; edge detection; image gradient; max distance;
Conference_Titel :
Image Processing, 2005. ICIP 2005. IEEE International Conference on
Print_ISBN :
0-7803-9134-9
DOI :
10.1109/ICIP.2005.1530596