A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications - PhD Dissertation

Aug 5, 2026

Unmanned Aerial Vehicles (UAVs) enable more capable and accessible remote data collection, including maintenance of electrical equipment across wide service areas. This dissertation addresses constraints in UAV cost, sensor payload weight/power, and platform compatibility by proposing a custom, relatively low-cost multispectral camera using accessible components for electrical arcing detection. Three industrial cameras capture visual, ultraviolet, and infrared bands, supporting arc-related emission and heat-driven infrared radiation. A custom dataset is generated in varied outdoor lighting, then machine learning classification models and ensemble approaches are trained and compared to evaluate strengths and limitations for low-cost sensor developers.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages152

Reading Time6 h 23 min

AccessFree to download

Abstract

UAV-based remote sensing and equipment maintenance

Custom low-cost multispectral camera system

Dataset generation in varied outdoor conditions

Machine learning model training and comparison

University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange | Doctoral Dissertations | Graduate School | | --- | --- | | 5-2025
A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications
Elizabeth Piersall
University of Tennessee, Knoxville, [epiersal@vols.utk.edu](epiersal@vols.utk.edu)
Follow this and additional works at: [https://trace.tennessee.edu/utk_graddiss](https://trace.tennessee.edu/utk_graddiss) | | Recommended Citation Piersall, Elizabeth, "A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications. " PhD diss., University of Tennessee, 2025. [https://trace.tennessee.edu/utk_graddiss/12406](https://trace.tennessee.edu/utk_graddiss/12406) This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [trace@utk.edu](trace@utk.edu). To the Graduate Council: I am submitting herewith a dissertation written by Elizabeth Piersall entitled "A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Data Science and Engineering. Peter L. Fuhr, Major Professor We have read this dissertation and recommend its acceptance: Yilu Liu, Fangxing F. Li , Ali R. Ekti Accepted for the Council: Dr. Amy Cathey Vice Provost and Dean of the Graduate School (Original signatures are on file with official student records.) To the Graduate Council: I am submitting herewith a dissertation written by Elizabeth Piersall entitled “A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications.” I have examined the final paper copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Data Science and Engineering. Dr. Peter Fuhr, Major Professor We have read this dissertation and recommend its acceptance: | Dr. Yilu Liu | | --- | | Dr. Fangxing Li | Dr. Ali Ekti Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School To the Graduate Council: I am submitting herewith a dissertation written by Elizabeth Piersall entitled “A Comparison of Machine Learning Based Arc Detection For a Custom Low-Cost Multispectral Camera for UAV Applications.” I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Data Science and Engineering. Dr. Peter Fuhr, Major Professor We have read this dissertation and recommend its acceptance: Dr. Yilu Liu Dr. Fangxing Li Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School (Original signatures are on file with official student records.) A Comparison of Machine Learning Based Arc Detection Fora Custom Low-Cost Multispectral Camera for UAV Applications A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Elizabeth Piersall May 2025 Abstract Unmanned Aerial Vehicles (UAVs), also known as drones, have seen increasingly widespread use in recent years with advancing technology in a variety of fields, and present an option to improve the capability and accessibility of remote data collection. One area that may benefit from this capacity is the m

FAQs

What problem does the dissertation address?

It addresses how to detect electrical arcing using a UAV-compatible, custom low-cost multispectral sensing system when conventional sensors can be expensive and platform-constrained.

How is the multispectral camera configured in this research?

It uses three industrial cameras collecting visual, ultraviolet, and infrared light, where arc emissions appear in visual/UV and nearby heating increases infrared radiation.

What methods are compared for arc detection?

The dissertation trains separate and ensemble machine learning image classification models, comparing model parameters and ensemble approaches to assess strengths and limitations.

Aug 5, 2026

Upload by River Wang