Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder

Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder

Pump maintenance plays a pivotal role in industrial operations, where timely detection of faults is key to avoiding costly downtimes. This research explores the influence of two audio feature extraction techniques, Mel-Frequency Cepstral Coefficients (MFCC) and Mel-spectrograms, on the effectiveness...

Full description

Bibliographic Details
Published in:Proceedings - 2024 2nd International Conference on Computer Graphics and Image Processing, CGIP 2024
Main Author: Bin Saharom A.S.; Ehara F.
Format: Conference paper
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2024
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195138022&doi=10.1109%2fCGIP62525.2024.00030&partnerID=40&md5=1fc5cf69559c98b2c3ada8c7fb5819b2
id 2-s2.0-85195138022
spelling 2-s2.0-85195138022
Bin Saharom A.S.; Ehara F.
Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
2024
Proceedings - 2024 2nd International Conference on Computer Graphics and Image Processing, CGIP 2024


10.1109/CGIP62525.2024.00030
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195138022&doi=10.1109%2fCGIP62525.2024.00030&partnerID=40&md5=1fc5cf69559c98b2c3ada8c7fb5819b2
Pump maintenance plays a pivotal role in industrial operations, where timely detection of faults is key to avoiding costly downtimes. This research explores the influence of two audio feature extraction techniques, Mel-Frequency Cepstral Coefficients (MFCC) and Mel-spectrograms, on the effectiveness of autoencoders in detecting pump faults. Using the Malfunctioning Industrial Machine Investigation and Inspection (MIMII) dataset, the study trained an auto encoder on normal pump sounds and evaluated it against a balanced test set of normal and anomalous sounds. The results present the superiority of Mel-spectrograms over MFCCs in various performance metrics. These findings emphasize the critical role of feature selection in autoencoder-based pump fault detection, marking a significant stride towards optimizing predictive maintenance strategies. © 2024 IEEE.
Institute of Electrical and Electronics Engineers Inc.

English
Conference paper
author Bin Saharom A.S.; Ehara F.
spellingShingle Bin Saharom A.S.; Ehara F.
Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
author_facet Bin Saharom A.S.; Ehara F.
author_sort Bin Saharom A.S.; Ehara F.
title Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
title_short Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
title_full Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
title_fullStr Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
title_full_unstemmed Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
title_sort Comparative Analysis of MFCC and Mel-Spectrogram Features in Pump Fault Detection Using Autoencoder
publishDate 2024
container_title Proceedings - 2024 2nd International Conference on Computer Graphics and Image Processing, CGIP 2024
container_volume
container_issue
doi_str_mv 10.1109/CGIP62525.2024.00030
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195138022&doi=10.1109%2fCGIP62525.2024.00030&partnerID=40&md5=1fc5cf69559c98b2c3ada8c7fb5819b2
description Pump maintenance plays a pivotal role in industrial operations, where timely detection of faults is key to avoiding costly downtimes. This research explores the influence of two audio feature extraction techniques, Mel-Frequency Cepstral Coefficients (MFCC) and Mel-spectrograms, on the effectiveness of autoencoders in detecting pump faults. Using the Malfunctioning Industrial Machine Investigation and Inspection (MIMII) dataset, the study trained an auto encoder on normal pump sounds and evaluated it against a balanced test set of normal and anomalous sounds. The results present the superiority of Mel-spectrograms over MFCCs in various performance metrics. These findings emphasize the critical role of feature selection in autoencoder-based pump fault detection, marking a significant stride towards optimizing predictive maintenance strategies. © 2024 IEEE.
publisher Institute of Electrical and Electronics Engineers Inc.
issn
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1809678013801955328

Similar Items