Feature transform for ATR image decomposition

Davi Geiger, Robert Hummel, Barney Baldwin, Tyng Luh Liu, Laxmi Parida

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

Abstract

We have developed an approach to image decomposition for ATR applications called the `feature transform.' There are two aspects to the feature transform: (1) A collection of rich, sophisticated feature extraction routines, and (2) the orchestration of a hierarchical decomposition of the scene into an image description based on the features. We have expanded the approach into two directions, one considering local features and the other considering global features. When studying local features, we have developed for (1) corner, T-junctions, edge, line, endstopping, and blob detectors as local features. A unified approach is used for all these detectors. For (2), we make use of the theory of matching pursuits and extend it to robust measures, using results involving L p norms, in order to build an iterative procedure in which local features are removed from the image successively, in a hierarchical manner. We have also considered for (1) global shape features or modal features, i.e., features representing the various modes of the models to be detected. For (2) a multiscale strategy is used for moving from the principal modes to secondary ones. The common aspect of both directions, local and global feature detection, is that the resulting transformations of the scene decomposes the image into a collection of features, in much the same way that a discrete Fourier transform decomposes an image into a sum of sinusoidal bar patterns. With the feature transform, however, the decomposition uses redundant basis functions that are related to spatially localized features or modal features that support the recognition process.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Pages512-523
Number of pages12
Volume2484
StatePublished - 1995
EventSignal Processing, Sensor Fusion, and Target Recognition IV - Orlando, FL, USA
Duration: Apr 17 1995Apr 19 1995

Other

OtherSignal Processing, Sensor Fusion, and Target Recognition IV
CityOrlando, FL, USA
Period4/17/954/19/95

Fingerprint

Decomposition
decomposition
Detectors
Discrete Fourier transforms
Feature extraction
detectors
norms
pattern recognition

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Geiger, D., Hummel, R., Baldwin, B., Liu, T. L., & Parida, L. (1995). Feature transform for ATR image decomposition. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 2484, pp. 512-523)

Feature transform for ATR image decomposition. / Geiger, Davi; Hummel, Robert; Baldwin, Barney; Liu, Tyng Luh; Parida, Laxmi.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 2484 1995. p. 512-523.

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

Geiger, D, Hummel, R, Baldwin, B, Liu, TL & Parida, L 1995, Feature transform for ATR image decomposition. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 2484, pp. 512-523, Signal Processing, Sensor Fusion, and Target Recognition IV, Orlando, FL, USA, 4/17/95.
Geiger D, Hummel R, Baldwin B, Liu TL, Parida L. Feature transform for ATR image decomposition. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 2484. 1995. p. 512-523
Geiger, Davi ; Hummel, Robert ; Baldwin, Barney ; Liu, Tyng Luh ; Parida, Laxmi. / Feature transform for ATR image decomposition. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 2484 1995. pp. 512-523
@inproceedings{762005bfbd30414a88fb84ddfb2ed288,
title = "Feature transform for ATR image decomposition",
abstract = "We have developed an approach to image decomposition for ATR applications called the `feature transform.' There are two aspects to the feature transform: (1) A collection of rich, sophisticated feature extraction routines, and (2) the orchestration of a hierarchical decomposition of the scene into an image description based on the features. We have expanded the approach into two directions, one considering local features and the other considering global features. When studying local features, we have developed for (1) corner, T-junctions, edge, line, endstopping, and blob detectors as local features. A unified approach is used for all these detectors. For (2), we make use of the theory of matching pursuits and extend it to robust measures, using results involving L p norms, in order to build an iterative procedure in which local features are removed from the image successively, in a hierarchical manner. We have also considered for (1) global shape features or modal features, i.e., features representing the various modes of the models to be detected. For (2) a multiscale strategy is used for moving from the principal modes to secondary ones. The common aspect of both directions, local and global feature detection, is that the resulting transformations of the scene decomposes the image into a collection of features, in much the same way that a discrete Fourier transform decomposes an image into a sum of sinusoidal bar patterns. With the feature transform, however, the decomposition uses redundant basis functions that are related to spatially localized features or modal features that support the recognition process.",
author = "Davi Geiger and Robert Hummel and Barney Baldwin and Liu, {Tyng Luh} and Laxmi Parida",
year = "1995",
language = "English (US)",
isbn = "0819418374",
volume = "2484",
pages = "512--523",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Feature transform for ATR image decomposition

AU - Geiger, Davi

AU - Hummel, Robert

AU - Baldwin, Barney

AU - Liu, Tyng Luh

AU - Parida, Laxmi

PY - 1995

Y1 - 1995

N2 - We have developed an approach to image decomposition for ATR applications called the `feature transform.' There are two aspects to the feature transform: (1) A collection of rich, sophisticated feature extraction routines, and (2) the orchestration of a hierarchical decomposition of the scene into an image description based on the features. We have expanded the approach into two directions, one considering local features and the other considering global features. When studying local features, we have developed for (1) corner, T-junctions, edge, line, endstopping, and blob detectors as local features. A unified approach is used for all these detectors. For (2), we make use of the theory of matching pursuits and extend it to robust measures, using results involving L p norms, in order to build an iterative procedure in which local features are removed from the image successively, in a hierarchical manner. We have also considered for (1) global shape features or modal features, i.e., features representing the various modes of the models to be detected. For (2) a multiscale strategy is used for moving from the principal modes to secondary ones. The common aspect of both directions, local and global feature detection, is that the resulting transformations of the scene decomposes the image into a collection of features, in much the same way that a discrete Fourier transform decomposes an image into a sum of sinusoidal bar patterns. With the feature transform, however, the decomposition uses redundant basis functions that are related to spatially localized features or modal features that support the recognition process.

AB - We have developed an approach to image decomposition for ATR applications called the `feature transform.' There are two aspects to the feature transform: (1) A collection of rich, sophisticated feature extraction routines, and (2) the orchestration of a hierarchical decomposition of the scene into an image description based on the features. We have expanded the approach into two directions, one considering local features and the other considering global features. When studying local features, we have developed for (1) corner, T-junctions, edge, line, endstopping, and blob detectors as local features. A unified approach is used for all these detectors. For (2), we make use of the theory of matching pursuits and extend it to robust measures, using results involving L p norms, in order to build an iterative procedure in which local features are removed from the image successively, in a hierarchical manner. We have also considered for (1) global shape features or modal features, i.e., features representing the various modes of the models to be detected. For (2) a multiscale strategy is used for moving from the principal modes to secondary ones. The common aspect of both directions, local and global feature detection, is that the resulting transformations of the scene decomposes the image into a collection of features, in much the same way that a discrete Fourier transform decomposes an image into a sum of sinusoidal bar patterns. With the feature transform, however, the decomposition uses redundant basis functions that are related to spatially localized features or modal features that support the recognition process.

UR - http://www.scopus.com/inward/record.url?scp=0029544538&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029544538&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0029544538

SN - 0819418374

SN - 9780819418371

VL - 2484

SP - 512

EP - 523

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

ER -