Ugochukwu Akpudo

Ugochukwu Akpudo

eXplainable AI | Research | Engineer

Publications

View all my articles on my Google Scholar profile.

Selected Journal Articles

Exploring the Efficiencies of Spectral Isolation for Intelligent Wear Monitoring of Micro Drill Bit Automatic Regrinding In-Line Systems

Published in Algorithms, 2022

This study explores artificial intelligence-based models for learning the discriminant spectral information stored in the vibration signals and considers the accuracy and cost implications of spectral isolation of the critical spectral segments of the signals for accurate equipment monitoring. Read more

Recommended citation: Akpudo, U.E.; Hur, J.-W. Exploring the Efficiencies of Spectral Isolation for Intelligent Wear Monitoring of Micro Drill Bit Automatic Regrinding In-Line Systems. Algorithms 2022, 15, 194. https://doi.org/10.3390/a15060194.
Download Paper

D-dCNN: A Novel Hybrid Deep Learning-Based Tool for Vibration-Based Diagnostics

Published in Energies, 2021

This study proposes a hybrid DNN and one-dimensional CNN diagnostics model (D-dCNN) which automatically extracts high-level discriminative features from vibration signals for fault detection and isolation (FDI). Read more

Recommended citation: Akpudo, U.E.; Hur, J.-W. D-dCNN: A Novel Hybrid Deep Learning-Based Tool for Vibration-Based Diagnostics. Energies 2021, 14, 5286. https://doi.org/10.3390/en14175286
Download Paper

An Explainable DL-Based Condition Monitoring Framework for Water-Emulsified Diesel CR Systems

Published in Electronics, 2021

This study proposes the use of common rail (CR) pressure differentials and a deep one-dimensional convolutional neural network (1D-CNN) with the local interpretable model-agnostic explanations (LIME) for empirical diagnostic evaluations (and validations) using a KIA Sorento 2004 four-cylinder line engine as a case study. Read more

Recommended citation: Akpudo, U.E.; Hur, J.-W. An Explainable DL-Based Condition Monitoring Framework for Water-Emulsified Diesel CR Systems. Electronics 2021, 10, 2522. https://doi.org/10.3390/electronics10202522.
Download Paper

An Automated Sensor Fusion Approach for the RUL Prediction of Electromagnetic Pumps

Published in IEEE Access, 2021

This study introduces a multi-sensor prognostics approach which merges highly prognosible statistical features from vibrational and pressure sensor measurements after a multi-level wavelet decomposition of the signals. Read more

Recommended citation: U. E. Akpudo and H. Jang-Wook, "An Automated Sensor Fusion Approach for the RUL Prediction of Electromagnetic Pumps," in IEEE Access, vol. 9, pp. 38920-38933, 2021, doi: 10.1109/ACCESS.2021.3063676.
Download Paper

Towards bearing failure prognostics: a practical comparison between data-driven methods for industrial applications

Published in Journal of Mechanical Science and Technology (JMST), 2020

This study presents a methodology for constructing a reliable HI for bearing prognostics, choosing a reliable TSP, and provides a comparison between ML and DL methods for bearing prognostics. Read more

Recommended citation: Akpudo, U.E., Hur, JW. Towards bearing failure prognostics: a practical comparison between data-driven methods for industrial applications. Journal of Mechanical Science and Technology (JMST) 34, 4161–4172 (2020). https://doi.org/10.1007/s12206-020-0908-7
Download Paper

A Cost-Efficient MFCC-Based Fault Detection and Isolation Technology for Electromagnetic Pumps

Published in IEEE Access, 2020

This study presents a robust approach for vibration-based failure diagnostics of electromagnetic/solenoid pumps which employ a multi-domain feature extraction procedure (statistical time-domain and frequency-domain features, Mel frequency cepstral coefficients, and continuous wavelet coefficients) for capturing linear and nonlinear properties from the signals. Read more

Recommended citation: U. E. Akpudo and H. Jang-Wook, "A Multi-Domain Diagnostics Approach for Solenoid Pumps Based on Discriminative Features," in IEEE Access, vol. 8, pp. 175020-175034, 2020, doi: 10.1109/ACCESS.2020.3025909..
Download Paper

Selected Conference Papers

Coherentice: Invertible Concept-Based Explainability Framework for CNNs beyond Fidelity

Published in 2024 IEEE International Conference on Multimedia and Expo (ICME), 2024

This paper extends the Invertible Concept-based Explainer (ICE) to introduce a new ingredient measuring concept consistency. Read more

Recommended citation: U. E. Akpudo, Y. Gao, J. Zhou and A. Lewis, "Coherentice: Invertible Concept-Based Explainability Framework for CNNs beyond Fidelity," 2024 IEEE International Conference on Multimedia and Expo (ICME), Niagara Falls, ON, Canada, 2024, pp. 1-6, doi: 10.1109/ICME57554.2024.10687699.
Download Paper | Download Slides

What EXACTLY are We Looking at?: Investigating for Discriminance in Ultra-Fine-Grained Visual Categorization Tasks

Published in 2023 International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2023

This study explores and draws for the first time, qualitative insights into Ultra-FGVC images through saliency-based explanation methods that provide intuitive hints on where the models are looking at the images. Read more

Recommended citation: U. E. Akpudo, X. Yu, J. Zhou and Y. Gao, "What EXACTLY are We Looking at?: Investigating for Discriminance in Ultra-Fine-Grained Visual Categorization Tasks," 2023 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Port Macquarie, Australia, 2023, pp. 129-136, doi: 10.1109/DICTA60407.2023.00026.
Download Paper

Towards an Inclusive Data Governance Policy for the use of AI in Africa

Published in Data for Policy 2022, 2022

This brief unveils some vulnerabilities surrounding the use of AI in SSA and promotes equitable access to new technologies in SSA amidst the anxiety around AI and concerns about data governance. Read more

Recommended citation: J. O. Effoduh, U. E. Akpudo, and J. D. Kong, “Toward a trustworthy and inclusive data governance policy for the use of artificial intelligence in Africa,” Data & Policy, vol. 6, p. e34, 2024. doi:10.1017/dap.2024.26.
Download Paper | Download Slides

[Best Paper]A Feature Fusion-based Prognostics Approach for Rolling Element Bearings

Published in 5th International Conference on Materials and Reliability Jeju, Korea, 2019

This short paper proposes a kernel principal component analysis (KPCA) feature fusion technique for degradation assessment, and a deep learning model for prognostics of rolling element bearings. Read more

Recommended citation: U. E. Akpudo and J. H. Hur, “A Feature Fusion-based Prognostics Approach for Rolling Element Bearings,” 5th International Conference on Materials and Reliability Jeju, Korea, Nov. 27-29, 2019.
Download Paper | Download Slides