Natural image understanding using algorithm selection and high-level feedback

Martin Lukac, Michitaka Kameyama, Kosuke Hiura

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

6 Citations (Scopus)

Abstract

Natural Image processing and understanding encompasses hundreds or even thousands of different algorithms. Each algorithm has a certain peak performance for a particular set of input features and configurations of the objects/regions of the input image (environment). To obtain the best possible result of processing, we propose an algorithm selection approach that permits to always use the most appropriate algorithm for the given input image. This is obtained by at first selecting an algorithm based on low level features such as color intensity, histograms, spectral coefficients. The resulting high level image description is then analyzed for logical inconsistencies (contradictions) that are then used to refine the selection of the processing elements. The feedback created from the contradiction information is executed by a Bayesian Network that integrates both the features and a higher level information selection processes. The selection stops when the high level inconsistencies are all resolved or no more different algorithms can be selected.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Volume8662
DOIs
Publication statusPublished - 2013
Externally publishedYes
EventIntelligent Robots and Computer Vision XXX: Algorithms and Techniques - Burlingame, CA, United States
Duration: Feb 4 2013Feb 6 2013

Other

OtherIntelligent Robots and Computer Vision XXX: Algorithms and Techniques
CountryUnited States
CityBurlingame, CA
Period2/4/132/6/13

Fingerprint

Image Understanding
Image understanding
Feedback
Inconsistency
Bayesian networks
Processing
Bayesian Networks
histograms
Histogram
image processing
Image Processing
Image processing
Integrate
Color
color
Configuration
Coefficient
coefficients
configurations

Keywords

  • Algorithm Selection
  • Bayesian Networks
  • Contour extraction
  • Image Processing

ASJC Scopus subject areas

  • Applied Mathematics
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

Lukac, M., Kameyama, M., & Hiura, K. (2013). Natural image understanding using algorithm selection and high-level feedback. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 8662). [86620D] https://doi.org/10.1117/12.2008593

Natural image understanding using algorithm selection and high-level feedback. / Lukac, Martin; Kameyama, Michitaka; Hiura, Kosuke.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 8662 2013. 86620D.

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

Lukac, M, Kameyama, M & Hiura, K 2013, Natural image understanding using algorithm selection and high-level feedback. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 8662, 86620D, Intelligent Robots and Computer Vision XXX: Algorithms and Techniques, Burlingame, CA, United States, 2/4/13. https://doi.org/10.1117/12.2008593
Lukac M, Kameyama M, Hiura K. Natural image understanding using algorithm selection and high-level feedback. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 8662. 2013. 86620D https://doi.org/10.1117/12.2008593
Lukac, Martin ; Kameyama, Michitaka ; Hiura, Kosuke. / Natural image understanding using algorithm selection and high-level feedback. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 8662 2013.
@inproceedings{ef833199eb95466a98080d171adc28da,
title = "Natural image understanding using algorithm selection and high-level feedback",
abstract = "Natural Image processing and understanding encompasses hundreds or even thousands of different algorithms. Each algorithm has a certain peak performance for a particular set of input features and configurations of the objects/regions of the input image (environment). To obtain the best possible result of processing, we propose an algorithm selection approach that permits to always use the most appropriate algorithm for the given input image. This is obtained by at first selecting an algorithm based on low level features such as color intensity, histograms, spectral coefficients. The resulting high level image description is then analyzed for logical inconsistencies (contradictions) that are then used to refine the selection of the processing elements. The feedback created from the contradiction information is executed by a Bayesian Network that integrates both the features and a higher level information selection processes. The selection stops when the high level inconsistencies are all resolved or no more different algorithms can be selected.",
keywords = "Algorithm Selection, Bayesian Networks, Contour extraction, Image Processing",
author = "Martin Lukac and Michitaka Kameyama and Kosuke Hiura",
year = "2013",
doi = "10.1117/12.2008593",
language = "English",
isbn = "9780819494351",
volume = "8662",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Natural image understanding using algorithm selection and high-level feedback

AU - Lukac, Martin

AU - Kameyama, Michitaka

AU - Hiura, Kosuke

PY - 2013

Y1 - 2013

N2 - Natural Image processing and understanding encompasses hundreds or even thousands of different algorithms. Each algorithm has a certain peak performance for a particular set of input features and configurations of the objects/regions of the input image (environment). To obtain the best possible result of processing, we propose an algorithm selection approach that permits to always use the most appropriate algorithm for the given input image. This is obtained by at first selecting an algorithm based on low level features such as color intensity, histograms, spectral coefficients. The resulting high level image description is then analyzed for logical inconsistencies (contradictions) that are then used to refine the selection of the processing elements. The feedback created from the contradiction information is executed by a Bayesian Network that integrates both the features and a higher level information selection processes. The selection stops when the high level inconsistencies are all resolved or no more different algorithms can be selected.

AB - Natural Image processing and understanding encompasses hundreds or even thousands of different algorithms. Each algorithm has a certain peak performance for a particular set of input features and configurations of the objects/regions of the input image (environment). To obtain the best possible result of processing, we propose an algorithm selection approach that permits to always use the most appropriate algorithm for the given input image. This is obtained by at first selecting an algorithm based on low level features such as color intensity, histograms, spectral coefficients. The resulting high level image description is then analyzed for logical inconsistencies (contradictions) that are then used to refine the selection of the processing elements. The feedback created from the contradiction information is executed by a Bayesian Network that integrates both the features and a higher level information selection processes. The selection stops when the high level inconsistencies are all resolved or no more different algorithms can be selected.

KW - Algorithm Selection

KW - Bayesian Networks

KW - Contour extraction

KW - Image Processing

UR - http://www.scopus.com/inward/record.url?scp=84875891053&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875891053&partnerID=8YFLogxK

U2 - 10.1117/12.2008593

DO - 10.1117/12.2008593

M3 - Conference contribution

AN - SCOPUS:84875891053

SN - 9780819494351

VL - 8662

BT - Proceedings of SPIE - The International Society for Optical Engineering

ER -