Discontinuous and Pattern Matching algorithm to measure deformation having discontinuities

Research output: Contribution to journalArticle

Abstract

Remote deformation monitoring with high accuracy is a challenging task. The technique of Digital Image Correction (DIC) is commonly used to measure remote deformation using images and provides high subpixel accuracy. However, DIC fails if deformation is non-continuous, which is a regularly occurring scenario in deformable solids due to the presence of discontinuities such as cracks and crevices. To overcome the limitation posed by DIC, a novel Discontinuous and Pattern Matching (DPM) algorithm is proposed in this study. Initially DPM algorithm demarcates the area where DIC fails by using the results of DIC. Later, DPM algorithm utilizes the features of pattern matching and embeds discontinuity in DIC to measure deformation in the demarcated areas where DIC failed. The performance of the proposed DPM algorithm is evaluated using two different experiments involving different types of discontinuities. The accuracy achieved in evaluation is higher than the normally required one-tenth of a pixel and the average absolute errors remained in the range of 0.02 to 0.07 pixels. The results are compared with another state-of-the-art DIC and pattern matching based technique and comparative analysis show that the proposed DPM algorithm improved the accuracy of deformation measurement in the range of 0.02 to 0.1 pixels depending on different scenarios.
Original languageEnglish
Pages (from-to)223
Number of pages1
JournalEngineering Applications of Artificial Intelligence
Volume81
DOIs
Publication statusPublished - May 2019

Fingerprint

Pattern matching
Pixels
Cracks
Monitoring

Cite this

@article{2b2577de8c5c44e3be262a8f11142d01,
title = "Discontinuous and Pattern Matching algorithm to measure deformation having discontinuities",
abstract = "Remote deformation monitoring with high accuracy is a challenging task. The technique of Digital Image Correction (DIC) is commonly used to measure remote deformation using images and provides high subpixel accuracy. However, DIC fails if deformation is non-continuous, which is a regularly occurring scenario in deformable solids due to the presence of discontinuities such as cracks and crevices. To overcome the limitation posed by DIC, a novel Discontinuous and Pattern Matching (DPM) algorithm is proposed in this study. Initially DPM algorithm demarcates the area where DIC fails by using the results of DIC. Later, DPM algorithm utilizes the features of pattern matching and embeds discontinuity in DIC to measure deformation in the demarcated areas where DIC failed. The performance of the proposed DPM algorithm is evaluated using two different experiments involving different types of discontinuities. The accuracy achieved in evaluation is higher than the normally required one-tenth of a pixel and the average absolute errors remained in the range of 0.02 to 0.07 pixels. The results are compared with another state-of-the-art DIC and pattern matching based technique and comparative analysis show that the proposed DPM algorithm improved the accuracy of deformation measurement in the range of 0.02 to 0.1 pixels depending on different scenarios.",
author = "Ghulam Hassan",
year = "2019",
month = "5",
doi = "10.1016/j.engappai.2019.02.017",
language = "English",
volume = "81",
pages = "223",
journal = "Engineering Applications of Artificial Intelligence",
issn = "0952-1976",
publisher = "Elsevier",

}

TY - JOUR

T1 - Discontinuous and Pattern Matching algorithm to measure deformation having discontinuities

AU - Hassan, Ghulam

PY - 2019/5

Y1 - 2019/5

N2 - Remote deformation monitoring with high accuracy is a challenging task. The technique of Digital Image Correction (DIC) is commonly used to measure remote deformation using images and provides high subpixel accuracy. However, DIC fails if deformation is non-continuous, which is a regularly occurring scenario in deformable solids due to the presence of discontinuities such as cracks and crevices. To overcome the limitation posed by DIC, a novel Discontinuous and Pattern Matching (DPM) algorithm is proposed in this study. Initially DPM algorithm demarcates the area where DIC fails by using the results of DIC. Later, DPM algorithm utilizes the features of pattern matching and embeds discontinuity in DIC to measure deformation in the demarcated areas where DIC failed. The performance of the proposed DPM algorithm is evaluated using two different experiments involving different types of discontinuities. The accuracy achieved in evaluation is higher than the normally required one-tenth of a pixel and the average absolute errors remained in the range of 0.02 to 0.07 pixels. The results are compared with another state-of-the-art DIC and pattern matching based technique and comparative analysis show that the proposed DPM algorithm improved the accuracy of deformation measurement in the range of 0.02 to 0.1 pixels depending on different scenarios.

AB - Remote deformation monitoring with high accuracy is a challenging task. The technique of Digital Image Correction (DIC) is commonly used to measure remote deformation using images and provides high subpixel accuracy. However, DIC fails if deformation is non-continuous, which is a regularly occurring scenario in deformable solids due to the presence of discontinuities such as cracks and crevices. To overcome the limitation posed by DIC, a novel Discontinuous and Pattern Matching (DPM) algorithm is proposed in this study. Initially DPM algorithm demarcates the area where DIC fails by using the results of DIC. Later, DPM algorithm utilizes the features of pattern matching and embeds discontinuity in DIC to measure deformation in the demarcated areas where DIC failed. The performance of the proposed DPM algorithm is evaluated using two different experiments involving different types of discontinuities. The accuracy achieved in evaluation is higher than the normally required one-tenth of a pixel and the average absolute errors remained in the range of 0.02 to 0.07 pixels. The results are compared with another state-of-the-art DIC and pattern matching based technique and comparative analysis show that the proposed DPM algorithm improved the accuracy of deformation measurement in the range of 0.02 to 0.1 pixels depending on different scenarios.

U2 - 10.1016/j.engappai.2019.02.017

DO - 10.1016/j.engappai.2019.02.017

M3 - Article

VL - 81

SP - 223

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

SN - 0952-1976

ER -