Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

An AI-Driven H5P Training for Epilepsy Emergency Management Preparedness in Dental Education.

European journal of dental education : official journal of the Association for Dental Education in Europe·2026
Same author

Shaping interprofessional identity and attitudes: scenario-based training in oral-systemic patient management.

BMC medical education·2025
Same author

Real-Time Anomaly Detection in Physiological Parameters: A Multi-Squad Monitoring and Communication Architecture.

Sensors (Basel, Switzerland)·2025
Same author

Introducing a specified on-line multimodal prehabilitation approach for total knee replacement surgery candidates using data from the COVID-19 pandemic: An exploratory field-based, pre-post, mixed methods implementation pilot study.

Journal of evaluation in clinical practice·2024
Same author

A holistic digital forensic analysis of Discord - Storage, memory, and network perspectives.

Journal of forensic sciences·2024
Same author

CANAttack: Assessing Vulnerabilities within Controller Area Network.

Sensors (Basel, Switzerland)·2023

Related Experiment Video

Updated: May 1, 2026

3D Printing of Preclinical X-ray Computed Tomographic Data Sets
11:06

3D Printing of Preclinical X-ray Computed Tomographic Data Sets

Published on: March 22, 2013

41.0K

3D printed firearm identification: A comparison of machine learning models.

Laura Garland1, Erasmus Mfodwo1, Ashar Neyaz2

  • 1Department of Computer Science, Sam Houston State University, Huntsville, Texas, USA.

Journal of Forensic Sciences
|April 28, 2026
PubMed
Summary

This study introduces a novel method for detecting 3D printed firearms by analyzing digital evidence. Machine learning models achieved 95.80% accuracy in classifying firearm objects from g-code files.

Keywords:
3D printingfirearmsg‐codemachine learning

More Related Videos

3D Printing of Biomolecular Models for Research and Pedagogy
09:17

3D Printing of Biomolecular Models for Research and Pedagogy

Published on: March 13, 2017

26.6K
Interactive Molecular Model Assembly with 3D Printing
06:15

Interactive Molecular Model Assembly with 3D Printing

Published on: August 13, 2020

11.6K

Related Experiment Videos

Last Updated: May 1, 2026

3D Printing of Preclinical X-ray Computed Tomographic Data Sets
11:06

3D Printing of Preclinical X-ray Computed Tomographic Data Sets

Published on: March 22, 2013

41.0K
3D Printing of Biomolecular Models for Research and Pedagogy
09:17

3D Printing of Biomolecular Models for Research and Pedagogy

Published on: March 13, 2017

26.6K
Interactive Molecular Model Assembly with 3D Printing
06:15

Interactive Molecular Model Assembly with 3D Printing

Published on: August 13, 2020

11.6K

Area of Science:

  • Digital Forensics
  • Computer Science
  • Mechanical Engineering

Background:

  • The proliferation of untraceable 3D printed firearms poses a significant security concern.
  • Current investigations heavily rely on analyzing physical objects, neglecting digital evidence from 3D printing instructions.

Purpose of the Study:

  • To develop and evaluate a machine learning-based approach for classifying 3D objects as firearms or non-firearms using g-code files.
  • To compare the effectiveness of different feature extraction methods and machine learning classifiers for digital firearm forensics.

Main Methods:

  • Extracted geometric features from g-code files using two methods: direct g-code analysis and 3D mesh construction.
  • Employed machine learning classifiers including Random Forest (RF), Support Vector Machine, Decision Tree, and Convolutional Neural Network.
  • Validated classification accuracy using 10-fold cross-validation.

Main Results:

  • The Random Forest (RF) model combined with the mesh construction method achieved the highest classification accuracy of 95.80%.
  • The mesh construction method consistently outperformed the direct g-code method in accuracy.
  • Statistical analysis confirmed the significance of the performance differences between methods.

Conclusions:

  • Digital evidence analysis of g-code files is a viable and accurate method for identifying 3D printed firearms.
  • The mesh construction feature extraction method offers superior performance for this classification task.
  • This approach provides a promising tool for law enforcement in combating untraceable firearms.