Detecting malicious defects in 3D printing process using machine learning and image classification

Mingtao Wu, Vir Phoha, Young Bai Moon, Amith K. Belman

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

7 Citations (Scopus)

Abstract

3D printing, or additive manufacturing, is a key technology for future manufacturing systems. However, 3D printing systems have unique vulnerabilities presented by the ability to affect the infill without affecting the exterior. In order to detect malicious infill defects in 3D printing process, this paper proposes the following: 1) investigate malicious defects in the 3D printing process, 2) extract features based on simulated 3D printing process images, and 3) an experiment of image classification with one group of non-defect infill image and the other group of defect infill training image from 3D printing process. The images are captured layer by layer from the top view of software simulation preview. The data extracted from images is input to two machine learning algorithms, Naive Bayes Classifier and J48 Decision Trees. The result shows Naive Bayes Classifier has an accuracy of 85.26% and J48 Decision Trees has an accuracy of 95.51% for classification.

Original languageEnglish (US)
Title of host publicationEmerging Technologies; Materials
Subtitle of host publicationGenetics to Structures; Safety Engineering and Risk Analysis
PublisherAmerican Society of Mechanical Engineers (ASME)
Volume14
ISBN (Electronic)9780791850688
DOIs
StatePublished - 2016
EventASME 2016 International Mechanical Engineering Congress and Exposition, IMECE 2016 - Phoenix, United States
Duration: Nov 11 2016Nov 17 2016

Other

OtherASME 2016 International Mechanical Engineering Congress and Exposition, IMECE 2016
CountryUnited States
CityPhoenix
Period11/11/1611/17/16

Fingerprint

Image classification
Learning systems
Printing
Defects
Decision trees
Classifiers
3D printers
Learning algorithms
Experiments

ASJC Scopus subject areas

  • Mechanical Engineering

Cite this

Wu, M., Phoha, V., Moon, Y. B., & Belman, A. K. (2016). Detecting malicious defects in 3D printing process using machine learning and image classification. In Emerging Technologies; Materials: Genetics to Structures; Safety Engineering and Risk Analysis (Vol. 14). American Society of Mechanical Engineers (ASME). https://doi.org/10.1115/IMECE201667641

Detecting malicious defects in 3D printing process using machine learning and image classification. / Wu, Mingtao; Phoha, Vir; Moon, Young Bai; Belman, Amith K.

Emerging Technologies; Materials: Genetics to Structures; Safety Engineering and Risk Analysis. Vol. 14 American Society of Mechanical Engineers (ASME), 2016.

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

Wu, M, Phoha, V, Moon, YB & Belman, AK 2016, Detecting malicious defects in 3D printing process using machine learning and image classification. in Emerging Technologies; Materials: Genetics to Structures; Safety Engineering and Risk Analysis. vol. 14, American Society of Mechanical Engineers (ASME), ASME 2016 International Mechanical Engineering Congress and Exposition, IMECE 2016, Phoenix, United States, 11/11/16. https://doi.org/10.1115/IMECE201667641
Wu M, Phoha V, Moon YB, Belman AK. Detecting malicious defects in 3D printing process using machine learning and image classification. In Emerging Technologies; Materials: Genetics to Structures; Safety Engineering and Risk Analysis. Vol. 14. American Society of Mechanical Engineers (ASME). 2016 https://doi.org/10.1115/IMECE201667641
Wu, Mingtao ; Phoha, Vir ; Moon, Young Bai ; Belman, Amith K. / Detecting malicious defects in 3D printing process using machine learning and image classification. Emerging Technologies; Materials: Genetics to Structures; Safety Engineering and Risk Analysis. Vol. 14 American Society of Mechanical Engineers (ASME), 2016.
@inproceedings{ea8148dd01ff4405b386e5d43f89fdae,
title = "Detecting malicious defects in 3D printing process using machine learning and image classification",
abstract = "3D printing, or additive manufacturing, is a key technology for future manufacturing systems. However, 3D printing systems have unique vulnerabilities presented by the ability to affect the infill without affecting the exterior. In order to detect malicious infill defects in 3D printing process, this paper proposes the following: 1) investigate malicious defects in the 3D printing process, 2) extract features based on simulated 3D printing process images, and 3) an experiment of image classification with one group of non-defect infill image and the other group of defect infill training image from 3D printing process. The images are captured layer by layer from the top view of software simulation preview. The data extracted from images is input to two machine learning algorithms, Naive Bayes Classifier and J48 Decision Trees. The result shows Naive Bayes Classifier has an accuracy of 85.26{\%} and J48 Decision Trees has an accuracy of 95.51{\%} for classification.",
author = "Mingtao Wu and Vir Phoha and Moon, {Young Bai} and Belman, {Amith K.}",
year = "2016",
doi = "10.1115/IMECE201667641",
language = "English (US)",
volume = "14",
booktitle = "Emerging Technologies; Materials",
publisher = "American Society of Mechanical Engineers (ASME)",

}

TY - GEN

T1 - Detecting malicious defects in 3D printing process using machine learning and image classification

AU - Wu, Mingtao

AU - Phoha, Vir

AU - Moon, Young Bai

AU - Belman, Amith K.

PY - 2016

Y1 - 2016

N2 - 3D printing, or additive manufacturing, is a key technology for future manufacturing systems. However, 3D printing systems have unique vulnerabilities presented by the ability to affect the infill without affecting the exterior. In order to detect malicious infill defects in 3D printing process, this paper proposes the following: 1) investigate malicious defects in the 3D printing process, 2) extract features based on simulated 3D printing process images, and 3) an experiment of image classification with one group of non-defect infill image and the other group of defect infill training image from 3D printing process. The images are captured layer by layer from the top view of software simulation preview. The data extracted from images is input to two machine learning algorithms, Naive Bayes Classifier and J48 Decision Trees. The result shows Naive Bayes Classifier has an accuracy of 85.26% and J48 Decision Trees has an accuracy of 95.51% for classification.

AB - 3D printing, or additive manufacturing, is a key technology for future manufacturing systems. However, 3D printing systems have unique vulnerabilities presented by the ability to affect the infill without affecting the exterior. In order to detect malicious infill defects in 3D printing process, this paper proposes the following: 1) investigate malicious defects in the 3D printing process, 2) extract features based on simulated 3D printing process images, and 3) an experiment of image classification with one group of non-defect infill image and the other group of defect infill training image from 3D printing process. The images are captured layer by layer from the top view of software simulation preview. The data extracted from images is input to two machine learning algorithms, Naive Bayes Classifier and J48 Decision Trees. The result shows Naive Bayes Classifier has an accuracy of 85.26% and J48 Decision Trees has an accuracy of 95.51% for classification.

UR - http://www.scopus.com/inward/record.url?scp=85021629295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021629295&partnerID=8YFLogxK

U2 - 10.1115/IMECE201667641

DO - 10.1115/IMECE201667641

M3 - Conference contribution

VL - 14

BT - Emerging Technologies; Materials

PB - American Society of Mechanical Engineers (ASME)

ER -