AMoEBA image segmentation: Modeling of individual voronoi tessellations

Nathan Johnson, Balu Karthikeyan, Daniel A. Ashlock, Kenneth M. Bryden

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Recent advancements in digital imaging systems have provided means for greater definition in photography. These technologies have brought high-resolution pictures that show greater clarity and allow for a 'deeper' analysis of information. This occurs through decreased distortion that may have otherwise resulted in features becoming 'blurred' or 'hidden'. However, this strength is accompanied by a weakness due to the large amount of data needed to display the otherwise non-apparent information; in turn creating massive file sizes which produce troubling or infeasible computation times during image storage and transfer. Therefore methods for handling these giant image files could be of great use in image processing and compression if an adequate amount of information can be maintained. During analysis the degree to which information must be replicated exactly depends on the purpose of the final product and the capability of computational resources. This work uses Graph Based Evolutionary Algorithms (GBEAs) to segment images into balanced-weight Voronoi tessellations which are then modeled by least squares surface fitting techniques using the Adaptive Modeling by Evolving Blocks Algorithm (AMoEBA). This twice optimized process yields new techniques of preserving image quality during image analysis and complex decision making. Methods in this paper begin by breaking an image into numerous components using balanced-weight Voronoi tessellations that are optimized to conform to features and detail in an image. These features are represented by surfaces through the strength and efficiency of the AMoEBA algorithm to find optimal representations that can differ from one tessellation to another.

Original languageEnglish (US)
Title of host publication2006 IEEE Congress on Evolutionary Computation, CEC 2006
Pages2127-2133
Number of pages7
StatePublished - 2006
Externally publishedYes
Event2006 IEEE Congress on Evolutionary Computation, CEC 2006 - Vancouver, BC, Canada
Duration: Jul 16 2006Jul 21 2006

Other

Other2006 IEEE Congress on Evolutionary Computation, CEC 2006
CountryCanada
CityVancouver, BC
Period7/16/067/21/06

Fingerprint

Voronoi Tessellation
Block Algorithm
Image segmentation
Image Segmentation
Modeling
Photography
Image compression
Evolutionary algorithms
Imaging systems
Image analysis
Image quality
Image processing
Surface Fitting
Digital Imaging
Decision making
Tessellation
Image Compression
Image Analysis
Imaging System
Image Quality

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Theoretical Computer Science

Cite this

Johnson, N., Karthikeyan, B., Ashlock, D. A., & Bryden, K. M. (2006). AMoEBA image segmentation: Modeling of individual voronoi tessellations. In 2006 IEEE Congress on Evolutionary Computation, CEC 2006 (pp. 2127-2133). [1688569]

AMoEBA image segmentation : Modeling of individual voronoi tessellations. / Johnson, Nathan; Karthikeyan, Balu; Ashlock, Daniel A.; Bryden, Kenneth M.

2006 IEEE Congress on Evolutionary Computation, CEC 2006. 2006. p. 2127-2133 1688569.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Johnson, N, Karthikeyan, B, Ashlock, DA & Bryden, KM 2006, AMoEBA image segmentation: Modeling of individual voronoi tessellations. in 2006 IEEE Congress on Evolutionary Computation, CEC 2006., 1688569, pp. 2127-2133, 2006 IEEE Congress on Evolutionary Computation, CEC 2006, Vancouver, BC, Canada, 7/16/06.
Johnson N, Karthikeyan B, Ashlock DA, Bryden KM. AMoEBA image segmentation: Modeling of individual voronoi tessellations. In 2006 IEEE Congress on Evolutionary Computation, CEC 2006. 2006. p. 2127-2133. 1688569
Johnson, Nathan ; Karthikeyan, Balu ; Ashlock, Daniel A. ; Bryden, Kenneth M. / AMoEBA image segmentation : Modeling of individual voronoi tessellations. 2006 IEEE Congress on Evolutionary Computation, CEC 2006. 2006. pp. 2127-2133
@inproceedings{6a7e884ded5449ffb9fd15def915966c,
title = "AMoEBA image segmentation: Modeling of individual voronoi tessellations",
abstract = "Recent advancements in digital imaging systems have provided means for greater definition in photography. These technologies have brought high-resolution pictures that show greater clarity and allow for a 'deeper' analysis of information. This occurs through decreased distortion that may have otherwise resulted in features becoming 'blurred' or 'hidden'. However, this strength is accompanied by a weakness due to the large amount of data needed to display the otherwise non-apparent information; in turn creating massive file sizes which produce troubling or infeasible computation times during image storage and transfer. Therefore methods for handling these giant image files could be of great use in image processing and compression if an adequate amount of information can be maintained. During analysis the degree to which information must be replicated exactly depends on the purpose of the final product and the capability of computational resources. This work uses Graph Based Evolutionary Algorithms (GBEAs) to segment images into balanced-weight Voronoi tessellations which are then modeled by least squares surface fitting techniques using the Adaptive Modeling by Evolving Blocks Algorithm (AMoEBA). This twice optimized process yields new techniques of preserving image quality during image analysis and complex decision making. Methods in this paper begin by breaking an image into numerous components using balanced-weight Voronoi tessellations that are optimized to conform to features and detail in an image. These features are represented by surfaces through the strength and efficiency of the AMoEBA algorithm to find optimal representations that can differ from one tessellation to another.",
author = "Nathan Johnson and Balu Karthikeyan and Ashlock, {Daniel A.} and Bryden, {Kenneth M.}",
year = "2006",
language = "English (US)",
isbn = "0780394879",
pages = "2127--2133",
booktitle = "2006 IEEE Congress on Evolutionary Computation, CEC 2006",

}

TY - GEN

T1 - AMoEBA image segmentation

T2 - Modeling of individual voronoi tessellations

AU - Johnson, Nathan

AU - Karthikeyan, Balu

AU - Ashlock, Daniel A.

AU - Bryden, Kenneth M.

PY - 2006

Y1 - 2006

N2 - Recent advancements in digital imaging systems have provided means for greater definition in photography. These technologies have brought high-resolution pictures that show greater clarity and allow for a 'deeper' analysis of information. This occurs through decreased distortion that may have otherwise resulted in features becoming 'blurred' or 'hidden'. However, this strength is accompanied by a weakness due to the large amount of data needed to display the otherwise non-apparent information; in turn creating massive file sizes which produce troubling or infeasible computation times during image storage and transfer. Therefore methods for handling these giant image files could be of great use in image processing and compression if an adequate amount of information can be maintained. During analysis the degree to which information must be replicated exactly depends on the purpose of the final product and the capability of computational resources. This work uses Graph Based Evolutionary Algorithms (GBEAs) to segment images into balanced-weight Voronoi tessellations which are then modeled by least squares surface fitting techniques using the Adaptive Modeling by Evolving Blocks Algorithm (AMoEBA). This twice optimized process yields new techniques of preserving image quality during image analysis and complex decision making. Methods in this paper begin by breaking an image into numerous components using balanced-weight Voronoi tessellations that are optimized to conform to features and detail in an image. These features are represented by surfaces through the strength and efficiency of the AMoEBA algorithm to find optimal representations that can differ from one tessellation to another.

AB - Recent advancements in digital imaging systems have provided means for greater definition in photography. These technologies have brought high-resolution pictures that show greater clarity and allow for a 'deeper' analysis of information. This occurs through decreased distortion that may have otherwise resulted in features becoming 'blurred' or 'hidden'. However, this strength is accompanied by a weakness due to the large amount of data needed to display the otherwise non-apparent information; in turn creating massive file sizes which produce troubling or infeasible computation times during image storage and transfer. Therefore methods for handling these giant image files could be of great use in image processing and compression if an adequate amount of information can be maintained. During analysis the degree to which information must be replicated exactly depends on the purpose of the final product and the capability of computational resources. This work uses Graph Based Evolutionary Algorithms (GBEAs) to segment images into balanced-weight Voronoi tessellations which are then modeled by least squares surface fitting techniques using the Adaptive Modeling by Evolving Blocks Algorithm (AMoEBA). This twice optimized process yields new techniques of preserving image quality during image analysis and complex decision making. Methods in this paper begin by breaking an image into numerous components using balanced-weight Voronoi tessellations that are optimized to conform to features and detail in an image. These features are represented by surfaces through the strength and efficiency of the AMoEBA algorithm to find optimal representations that can differ from one tessellation to another.

UR - http://www.scopus.com/inward/record.url?scp=34547372259&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547372259&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:34547372259

SN - 0780394879

SN - 9780780394872

SP - 2127

EP - 2133

BT - 2006 IEEE Congress on Evolutionary Computation, CEC 2006

ER -