Color image enhancement using a multiple-scale opponent neural network

Tian Pu; Jie Zhang; Guoqiang Ni

doi:10.1117/1.1787531

1 October 2004 Color image enhancement using a multiple-scale opponent neural network

Tian Pu, Jie Zhang, Guoqiang Ni

Author Affiliations +

Optical Engineering, Vol. 43, Issue 10, (October 2004). https://doi.org/10.1117/1.1787531

Abstract

One goal of image processing is to recreate the human visual perception of the original scene. We offer a method of color image enhancement for realistic image display. The algorithm is based on a special implementation of the feed-forward ON-OFF shunting neural network. The properties of the neural network for image processing are first investigated. The computational simulations make clear the strengths and weaknesses of the neural network. Combined with a brightness-based tone reproduction operator, a multiscale extension to the neural network is developed to correct the original version's weaknesses and provides a coherent framework for color image enhancement. The algorithm achieves both dynamic range compression and vivid color rendition and qualitatively approaches the aspects of early visual perception. Extensive tests demonstrate that this method could produce satisfying images without any unexpected results.

©(2004) Society of Photo-Optical Instrumentation Engineers (SPIE)

Citation Download Citation

Tian Pu, Jie Zhang, and Guoqiang Ni "Color image enhancement using a multiple-scale opponent neural network," Optical Engineering 43(10), (1 October 2004). https://doi.org/10.1117/1.1787531

Published: 1 October 2004

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