Image Halftoning with Cellular Neural Networks

Kenneth R. Crounse, Leon O. Chua, T. Roska, Tamás Roska

Research output: Contribution to journalArticle

57 Citations (Scopus)

Abstract

Many algorithms have been devised for halftoning digital images. These algorithms all suffer well-studied defects, which are especially apparent in the case where the resulting image is displayed at the marginally oversampled resolution and viewed at the critical pixel merge distance. Recently, it has been shown that a neural network approach may be useful for halftoning. Here, the feasibility of using neural networks in a practical application is considered. The cellular neural network (CNN) architecture is chosen for its proven implementability in VLSI and high speed operation. Since both the CNN and halftoning have a geometrically local character, the CNN provides a natural implementation. The CNN template weights are derived by analogy to the well-known error diffusion algorithm for halftoning. Some limitations of the neural network approach are analyzed providing an advance in designing template weights over previous methods. These limitations are shown to be especially critical in the case of the small interconnection neighborhoods needed for efficient implementation. Our design criteria are validated by direct simulation. The resulting halftones are shown to be more faithful reproductions of the original than those pro duced by the error diffusion algorithm. It is suggested that a CNN with optical inputs could provide a high-speed scanner/halftoner for applications such as FAX.

Original languageEnglish
Pages (from-to)267-283
Number of pages17
JournalIEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing
Volume40
Issue number4
DOIs
Publication statusPublished - 1993

Fingerprint

Cellular neural networks
Neural networks
Network architecture
Pixels
Defects

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing

Cite this

Image Halftoning with Cellular Neural Networks. / Crounse, Kenneth R.; Chua, Leon O.; Roska, T.; Roska, Tamás.

In: IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, Vol. 40, No. 4, 1993, p. 267-283.

Research output: Contribution to journalArticle

Crounse, Kenneth R. ; Chua, Leon O. ; Roska, T. ; Roska, Tamás. / Image Halftoning with Cellular Neural Networks. In: IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing. 1993 ; Vol. 40, No. 4. pp. 267-283.
@article{f557f255e216472797c80a1d4ab16fc3,
title = "Image Halftoning with Cellular Neural Networks",
abstract = "Many algorithms have been devised for halftoning digital images. These algorithms all suffer well-studied defects, which are especially apparent in the case where the resulting image is displayed at the marginally oversampled resolution and viewed at the critical pixel merge distance. Recently, it has been shown that a neural network approach may be useful for halftoning. Here, the feasibility of using neural networks in a practical application is considered. The cellular neural network (CNN) architecture is chosen for its proven implementability in VLSI and high speed operation. Since both the CNN and halftoning have a geometrically local character, the CNN provides a natural implementation. The CNN template weights are derived by analogy to the well-known error diffusion algorithm for halftoning. Some limitations of the neural network approach are analyzed providing an advance in designing template weights over previous methods. These limitations are shown to be especially critical in the case of the small interconnection neighborhoods needed for efficient implementation. Our design criteria are validated by direct simulation. The resulting halftones are shown to be more faithful reproductions of the original than those pro duced by the error diffusion algorithm. It is suggested that a CNN with optical inputs could provide a high-speed scanner/halftoner for applications such as FAX.",
author = "Crounse, {Kenneth R.} and Chua, {Leon O.} and T. Roska and Tam{\'a}s Roska",
year = "1993",
doi = "10.1109/82.224318",
language = "English",
volume = "40",
pages = "267--283",
journal = "IEEE Transactions on Circuits and Systems II: Express Briefs",
issn = "1057-7130",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Image Halftoning with Cellular Neural Networks

AU - Crounse, Kenneth R.

AU - Chua, Leon O.

AU - Roska, T.

AU - Roska, Tamás

PY - 1993

Y1 - 1993

N2 - Many algorithms have been devised for halftoning digital images. These algorithms all suffer well-studied defects, which are especially apparent in the case where the resulting image is displayed at the marginally oversampled resolution and viewed at the critical pixel merge distance. Recently, it has been shown that a neural network approach may be useful for halftoning. Here, the feasibility of using neural networks in a practical application is considered. The cellular neural network (CNN) architecture is chosen for its proven implementability in VLSI and high speed operation. Since both the CNN and halftoning have a geometrically local character, the CNN provides a natural implementation. The CNN template weights are derived by analogy to the well-known error diffusion algorithm for halftoning. Some limitations of the neural network approach are analyzed providing an advance in designing template weights over previous methods. These limitations are shown to be especially critical in the case of the small interconnection neighborhoods needed for efficient implementation. Our design criteria are validated by direct simulation. The resulting halftones are shown to be more faithful reproductions of the original than those pro duced by the error diffusion algorithm. It is suggested that a CNN with optical inputs could provide a high-speed scanner/halftoner for applications such as FAX.

AB - Many algorithms have been devised for halftoning digital images. These algorithms all suffer well-studied defects, which are especially apparent in the case where the resulting image is displayed at the marginally oversampled resolution and viewed at the critical pixel merge distance. Recently, it has been shown that a neural network approach may be useful for halftoning. Here, the feasibility of using neural networks in a practical application is considered. The cellular neural network (CNN) architecture is chosen for its proven implementability in VLSI and high speed operation. Since both the CNN and halftoning have a geometrically local character, the CNN provides a natural implementation. The CNN template weights are derived by analogy to the well-known error diffusion algorithm for halftoning. Some limitations of the neural network approach are analyzed providing an advance in designing template weights over previous methods. These limitations are shown to be especially critical in the case of the small interconnection neighborhoods needed for efficient implementation. Our design criteria are validated by direct simulation. The resulting halftones are shown to be more faithful reproductions of the original than those pro duced by the error diffusion algorithm. It is suggested that a CNN with optical inputs could provide a high-speed scanner/halftoner for applications such as FAX.

UR - http://www.scopus.com/inward/record.url?scp=0027578080&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027578080&partnerID=8YFLogxK

U2 - 10.1109/82.224318

DO - 10.1109/82.224318

M3 - Article

VL - 40

SP - 267

EP - 283

JO - IEEE Transactions on Circuits and Systems II: Express Briefs

JF - IEEE Transactions on Circuits and Systems II: Express Briefs

SN - 1057-7130

IS - 4

ER -