Rule Based Composite Gradient Edge Operator

Ron Patton; Dragana Brzakovic

doi:10.1117/12.940542

16 September 1987 Rule Based Composite Gradient Edge Operator

Ron Patton, Dragana Brzakovic

Proceedings Volume 0781, Infrared Image Processing and Enhancement; (1987) https://doi.org/10.1117/12.940542
Event: 1987 Technical Symposium Southeast on Optics, Electro-Optics, and Sensors, 1987, Orlando, FL, United States

Abstract

A gradient operator for detecting edges in imagery corrupted by noise is presented. Edge detectors that use masks to approximate derivative operations become imprecise in the presence of noise. Edge detectors which operate over small spatial scales (3 by 3) are ideal for edge localization and representing small detail; however, they respond equally well to high frequency noise and coarse textural information. Larger neighborhood operators (7 by 7) reduce the response to noise; however, edge localization and detail may be jeopardized. The Rule Based Composite Gradient Edge Operator (RBCGEO) interprets and integrates the responses of three operators each designed to extract edge information at a specific spatial resolution. The premise is that true edges have properties which will carry through several spatial scales while noise and other ambiguities will not. The RBCGEO uses a rule set to assign an edge magnitude, direction, and confidence of edge based on the composite response of the three operators. The algorithm and experimental results are presented.

Citation Download Citation

Ron Patton and Dragana Brzakovic "Rule Based Composite Gradient Edge Operator", Proc. SPIE 0781, Infrared Image Processing and Enhancement, (16 September 1987); https://doi.org/10.1117/12.940542

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available