Anomaly Detection in Metal-Textile Industries. Elsen, I., Ferrein, A., & Schiffer, S. In Anomaly Detection - Methods, Complexities and Applications, 0. IntechOpen, Rijeka, 2025. ![Anomaly Detection in Metal-Textile Industries [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex In this paper, we presented an approach to deploying a student–teacher feature pyramid model (STFPM) for anomaly detection metal-textile gas filters used in automotive exhaust gas filtering at GKD-Gebr. Kufferath AG. As the customer requires 100 percent quality of the delivered parts, an optical inspection process of every produced filter is required. This is very demanding for the human inspection worker as she has to inspect many 100 parts in an 8 hours shift. On the other hand, a fully vision-based system is not able to achieve the required classification rates either. Therefore, we propose a one-class anomaly detection process for the gas filters where human and AI work together in achieving the 100 percent pass rate. The STFPM model deals with the large amount of clearly true positive cases and automatedly classified them as PASS. Only cases of doubt where an anomaly has been detected are inspected by the human inspector. This way, the work load of the inspection worker is reduced, and, on the other hand, the hard to meet case of no mis-classification of the AI system can be avoided. We show the network architecture and the integration into the quality inspection process of the company GKD.
@incollection{ Elsen:Ferrein:Schiffer_InTech2025_AnomalyDetection,
author = {Ingo Elsen and Alexander Ferrein and Stefan Schiffer},
title = {Anomaly Detection in Metal-Textile Industries},
booktitle = {Anomaly Detection - Methods, Complexities and Applications},
publisher = {IntechOpen},
address = {Rijeka},
year = {2025},
editor = {Dr. Miguel Delgado-Prieto},
chapter = {0},
doi = {10.5772/intechopen.1008411},
url = {https://doi.org/10.5772/intechopen.1008411},
abstract = {In this paper, we presented an approach to deploying
a student–teacher feature pyramid model (STFPM) for
anomaly detection metal-textile gas filters used in
automotive exhaust gas filtering at
GKD-Gebr. Kufferath AG. As the customer requires 100
percent quality of the delivered parts, an optical
inspection process of every produced filter is
required. This is very demanding for the human
inspection worker as she has to inspect many 100
parts in an 8 hours shift. On the other hand, a
fully vision-based system is not able to achieve the
required classification rates either. Therefore, we
propose a one-class anomaly detection process for
the gas filters where human and AI work together in
achieving the 100 percent pass rate. The STFPM model
deals with the large amount of clearly true positive
cases and automatedly classified them as PASS. Only
cases of doubt where an anomaly has been detected
are inspected by the human inspector. This way, the
work load of the inspection worker is reduced, and,
on the other hand, the hard to meet case of no
mis-classification of the AI system can be
avoided. We show the network architecture and the
integration into the quality inspection process of
the company GKD.},
}
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 2024
% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Downloads: 0
{"_id":"4SE7u5YQ2cvrHbBNW","bibbaseid":"elsen-ferrein-schiffer-anomalydetectioninmetaltextileindustries-2025","author_short":["Elsen, I.","Ferrein, A.","Schiffer, S."],"bibdata":{"bibtype":"incollection","type":"incollection","author":[{"firstnames":["Ingo"],"propositions":[],"lastnames":["Elsen"],"suffixes":[]},{"firstnames":["Alexander"],"propositions":[],"lastnames":["Ferrein"],"suffixes":[]},{"firstnames":["Stefan"],"propositions":[],"lastnames":["Schiffer"],"suffixes":[]}],"title":"Anomaly Detection in Metal-Textile Industries","booktitle":"Anomaly Detection - Methods, Complexities and Applications","publisher":"IntechOpen","address":"Rijeka","year":"2025","editor":[{"firstnames":["Dr.","Miguel"],"propositions":[],"lastnames":["Delgado-Prieto"],"suffixes":[]}],"chapter":"0","doi":"10.5772/intechopen.1008411","url":"https://doi.org/10.5772/intechopen.1008411","abstract":"In this paper, we presented an approach to deploying a student–teacher feature pyramid model (STFPM) for anomaly detection metal-textile gas filters used in automotive exhaust gas filtering at GKD-Gebr. Kufferath AG. As the customer requires 100 percent quality of the delivered parts, an optical inspection process of every produced filter is required. This is very demanding for the human inspection worker as she has to inspect many 100 parts in an 8 hours shift. On the other hand, a fully vision-based system is not able to achieve the required classification rates either. Therefore, we propose a one-class anomaly detection process for the gas filters where human and AI work together in achieving the 100 percent pass rate. The STFPM model deals with the large amount of clearly true positive cases and automatedly classified them as PASS. Only cases of doubt where an anomaly has been detected are inspected by the human inspector. This way, the work load of the inspection worker is reduced, and, on the other hand, the hard to meet case of no mis-classification of the AI system can be avoided. We show the network architecture and the integration into the quality inspection process of the company GKD.","bibtex":"@incollection{ Elsen:Ferrein:Schiffer_InTech2025_AnomalyDetection,\n author = {Ingo Elsen and Alexander Ferrein and Stefan Schiffer},\n title = {Anomaly Detection in Metal-Textile Industries},\n booktitle = {Anomaly Detection - Methods, Complexities and Applications},\n publisher = {IntechOpen},\n address = {Rijeka},\n year = {2025},\n editor = {Dr. Miguel Delgado-Prieto},\n chapter = {0},\n doi = {10.5772/intechopen.1008411},\n url = {https://doi.org/10.5772/intechopen.1008411},\n abstract = {In this paper, we presented an approach to deploying\n a student–teacher feature pyramid model (STFPM) for\n anomaly detection metal-textile gas filters used in\n automotive exhaust gas filtering at\n GKD-Gebr. Kufferath AG. As the customer requires 100\n percent quality of the delivered parts, an optical\n inspection process of every produced filter is\n required. This is very demanding for the human\n inspection worker as she has to inspect many 100\n parts in an 8 hours shift. On the other hand, a\n fully vision-based system is not able to achieve the\n required classification rates either. Therefore, we\n propose a one-class anomaly detection process for\n the gas filters where human and AI work together in\n achieving the 100 percent pass rate. The STFPM model\n deals with the large amount of clearly true positive\n cases and automatedly classified them as PASS. Only\n cases of doubt where an anomaly has been detected\n are inspected by the human inspector. This way, the\n work load of the inspection worker is reduced, and,\n on the other hand, the hard to meet case of no\n mis-classification of the AI system can be\n avoided. We show the network architecture and the\n integration into the quality inspection process of\n the company GKD.},\n}\n% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n% 2024\n% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n\n","author_short":["Elsen, I.","Ferrein, A.","Schiffer, S."],"editor_short":["Delgado-Prieto, D. M."],"key":"Elsen:Ferrein:Schiffer_InTech2025_AnomalyDetection","id":"Elsen:Ferrein:Schiffer_InTech2025_AnomalyDetection","bibbaseid":"elsen-ferrein-schiffer-anomalydetectioninmetaltextileindustries-2025","role":"author","urls":{"Paper":"https://doi.org/10.5772/intechopen.1008411"},"metadata":{"authorlinks":{}}},"bibtype":"incollection","biburl":"http://maskor.fh-aachen.de/biblio/MASKOR.bib","dataSources":["i8ftsMK5wMbiNqmtB","pBDNm3knLemYTNMHw"],"keywords":[],"search_terms":["anomaly","detection","metal","textile","industries","elsen","ferrein","schiffer"],"title":"Anomaly Detection in Metal-Textile Industries","year":2025}