𝔖 Bobbio Scriptorium
✦   LIBER   ✦

A neural network-based nonlinear filter for image enhancement

✍ Scribed by S. Zhang; E. Salari

Publisher
John Wiley and Sons
Year
2002
Tongue
English
Weight
882 KB
Volume
12
Category
Article
ISSN
0899-9457

No coin nor oath required. For personal study only.

✦ Synopsis

Abstract

This paper explores a novel neural network‐based nonlinear filter that has the ability to remove mixed noises and sharpen the edges in noise‐corrupted digital images. The noise is assumed to be a mixture of both Gaussian and impulse types. Initially, a nonlinear filter is used to reduce the noise. The smoothed image is then combined with the output of an edge detector using a synthesizer to provide the effect of noise reducing and edge sharpening. The smoother and synthesizer are designed by using layered neural networks. Simulation results show that the proposed filter can effectively remove the mixed Guassian and impulsive noises and sharpen the edges. It can adapt itself to the various noise environments by learning during the training process. Β© 2002 Wiley Periodicals, Inc. Int J Imaging Syst Technol 12, 56–62, 2002; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10009

πŸ“œ SIMILAR VOLUMES

Convolution connection paradigm neural n
Convolution connection paradigm neural network enables linear system theory-based image enhancement
✍ William E. Blass; Stephen L. Mahan; Gordon Chin πŸ“‚ Article πŸ“… 1995 πŸ› John Wiley and Sons 🌐 English βš– 981 KB

A new type of robust numeric solution of one form of the Fredholm integral equations of the first kind has been discovered. This discovery has its most immediate and important applications in the deconvolution or deblurring of data acquired using scientific instruments. A solution of this integral e

Remote sensing image segmentation using
Remote sensing image segmentation using a Kalman filter-trained neural network
✍ K. S. Chen; D. H. Tsay; W. P. Huang; Y. C. Tzeng πŸ“‚ Article πŸ“… 1996 πŸ› John Wiley and Sons 🌐 English βš– 774 KB

This article describes the application of a neural network to the segmentation of remote sensing images of multispectral SPOT and fully polarimetric SAR data. The structure of the network is a modified multilayer perceptron and is trained by the Kalman filter theory. The internal activity of the net

An eight layer cellular neural network f
An eight layer cellular neural network for spatio-temporal image filtering
✍ Bertram E. Shi πŸ“‚ Article πŸ“… 2006 πŸ› John Wiley and Sons 🌐 English βš– 754 KB

## Abstract Spatio‐temporal filters are critical components of biologically inspired or neuromorphic algorithms for image motion analysis. In this paper, we describe eight layer cellular neural network architectures that can be used to implement these filters. Despite the apparently large number of