Symbolic dynamic filtering for image analysis: Theory and experimental validation

Aparna Subbu, Abhishek Srivastav, Asok Ray, Eric Keller

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Recent literature has reported the theory of symbolic dynamic filtering (SDF) of one-dimensional time-series data and its various applications for anomaly detection and pattern recognition. This paper extends the theory of SDF in the two-dimensional domain, where symbol sequences are generated from image data (i.e., pixels). Given the symbol sequence, a probabilistic finite state automaton (PFSA), called the D-Markov machine, is constructed on the principles of Markov random fields to incorporate the spatial information in the local neighborhoods of a pixel. The image analysis algorithm has been experimentally validated on a computer-controlled fatigue test apparatus that is equipped with a traveling optical microscope and ultrasonic flaw detectors. The surface images of test specimens, made of a polycrystalline alloy, are analyzed to detect and quantify the evolution of fatigue damage. The results of two-dimensional SDF analysis are in close agreement with those obtained from analysis of one-dimensional time-series data from the ultrasonic sensor, which are simultaneously generated from the same test specimen.

Original languageEnglish (US)
Pages (from-to)319-329
Number of pages11
JournalSignal, Image and Video Processing
Volume4
Issue number3
DOIs
StatePublished - Jan 1 2010

Fingerprint

Image analysis
Time series
Pixels
Ultrasonic sensors
Fatigue damage
Finite automata
Pattern recognition
Microscopes
Ultrasonics
Fatigue of materials
Detectors
Defects

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Subbu, Aparna ; Srivastav, Abhishek ; Ray, Asok ; Keller, Eric. / Symbolic dynamic filtering for image analysis : Theory and experimental validation. In: Signal, Image and Video Processing. 2010 ; Vol. 4, No. 3. pp. 319-329.
@article{9666188242ba4391b351db308296b9ea,
title = "Symbolic dynamic filtering for image analysis: Theory and experimental validation",
abstract = "Recent literature has reported the theory of symbolic dynamic filtering (SDF) of one-dimensional time-series data and its various applications for anomaly detection and pattern recognition. This paper extends the theory of SDF in the two-dimensional domain, where symbol sequences are generated from image data (i.e., pixels). Given the symbol sequence, a probabilistic finite state automaton (PFSA), called the D-Markov machine, is constructed on the principles of Markov random fields to incorporate the spatial information in the local neighborhoods of a pixel. The image analysis algorithm has been experimentally validated on a computer-controlled fatigue test apparatus that is equipped with a traveling optical microscope and ultrasonic flaw detectors. The surface images of test specimens, made of a polycrystalline alloy, are analyzed to detect and quantify the evolution of fatigue damage. The results of two-dimensional SDF analysis are in close agreement with those obtained from analysis of one-dimensional time-series data from the ultrasonic sensor, which are simultaneously generated from the same test specimen.",
author = "Aparna Subbu and Abhishek Srivastav and Asok Ray and Eric Keller",
year = "2010",
month = "1",
day = "1",
doi = "10.1007/s11760-009-0122-7",
language = "English (US)",
volume = "4",
pages = "319--329",
journal = "Signal, Image and Video Processing",
issn = "1863-1703",
publisher = "Springer London",
number = "3",

}

Symbolic dynamic filtering for image analysis : Theory and experimental validation. / Subbu, Aparna; Srivastav, Abhishek; Ray, Asok; Keller, Eric.

In: Signal, Image and Video Processing, Vol. 4, No. 3, 01.01.2010, p. 319-329.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Symbolic dynamic filtering for image analysis

T2 - Theory and experimental validation

AU - Subbu, Aparna

AU - Srivastav, Abhishek

AU - Ray, Asok

AU - Keller, Eric

PY - 2010/1/1

Y1 - 2010/1/1

N2 - Recent literature has reported the theory of symbolic dynamic filtering (SDF) of one-dimensional time-series data and its various applications for anomaly detection and pattern recognition. This paper extends the theory of SDF in the two-dimensional domain, where symbol sequences are generated from image data (i.e., pixels). Given the symbol sequence, a probabilistic finite state automaton (PFSA), called the D-Markov machine, is constructed on the principles of Markov random fields to incorporate the spatial information in the local neighborhoods of a pixel. The image analysis algorithm has been experimentally validated on a computer-controlled fatigue test apparatus that is equipped with a traveling optical microscope and ultrasonic flaw detectors. The surface images of test specimens, made of a polycrystalline alloy, are analyzed to detect and quantify the evolution of fatigue damage. The results of two-dimensional SDF analysis are in close agreement with those obtained from analysis of one-dimensional time-series data from the ultrasonic sensor, which are simultaneously generated from the same test specimen.

AB - Recent literature has reported the theory of symbolic dynamic filtering (SDF) of one-dimensional time-series data and its various applications for anomaly detection and pattern recognition. This paper extends the theory of SDF in the two-dimensional domain, where symbol sequences are generated from image data (i.e., pixels). Given the symbol sequence, a probabilistic finite state automaton (PFSA), called the D-Markov machine, is constructed on the principles of Markov random fields to incorporate the spatial information in the local neighborhoods of a pixel. The image analysis algorithm has been experimentally validated on a computer-controlled fatigue test apparatus that is equipped with a traveling optical microscope and ultrasonic flaw detectors. The surface images of test specimens, made of a polycrystalline alloy, are analyzed to detect and quantify the evolution of fatigue damage. The results of two-dimensional SDF analysis are in close agreement with those obtained from analysis of one-dimensional time-series data from the ultrasonic sensor, which are simultaneously generated from the same test specimen.

UR - http://www.scopus.com/inward/record.url?scp=77955848238&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955848238&partnerID=8YFLogxK

U2 - 10.1007/s11760-009-0122-7

DO - 10.1007/s11760-009-0122-7

M3 - Article

AN - SCOPUS:77955848238

VL - 4

SP - 319

EP - 329

JO - Signal, Image and Video Processing

JF - Signal, Image and Video Processing

SN - 1863-1703

IS - 3

ER -