Slide
Centre Interdisciplinaire
de Recherche et d’Innovation
en Cybersécurité et Société
de Recherche et d’Innovation
en Cybersécurité et Société
1.
Amirkhani, D.; Allili, M. S.; Hebbache, L.; Hammouche, N.; Lapointe, J.
Visual Concrete Bridge Defect Classification and Detection Using Deep Learning: A Systematic Review Article de journal
Dans: IEEE Transactions on Intelligent Transportation Systems, p. 1–23, 2024, ISSN: 15249050, (Publisher: Institute of Electrical and Electronics Engineers Inc.).
Résumé | Liens | BibTeX | Étiquettes: Annotation, Annotations, Bridges, Classification, Concrete, Concrete bridge defect, Concrete bridge defects, Concrete bridges, Concrete defects, Concretes, Deep learning, Defect classification, Defect detection, Defects, Detection, Inspection, Reviews, Segmentation, Taxonomies, Visualization
@article{amirkhani_visual_2024,
title = {Visual Concrete Bridge Defect Classification and Detection Using Deep Learning: A Systematic Review},
author = {D. Amirkhani and M. S. Allili and L. Hebbache and N. Hammouche and J. Lapointe},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85186994244&doi=10.1109%2fTITS.2024.3365296&partnerID=40&md5=a9228252d620ad6d444cc395296ebac2},
doi = {10.1109/TITS.2024.3365296},
issn = {15249050},
year = {2024},
date = {2024-01-01},
journal = {IEEE Transactions on Intelligent Transportation Systems},
pages = {1–23},
abstract = {Visual inspection is an important process for maintaining bridges in road transportation systems, and preventing catastrophic events and tragedies. In this process, accurate and automatic concrete defect classification and detection are major components to ensure early identification of any issue that can compromise the bridge safety and integrity. While a tremendous body of research has been proposed in the last decades for addressing these problems, the advent of deep learning unleashed huge opportunities for building more accurate and efficient methods. Our aim in this survey is to study the recent progress of vision-based concrete bridge defect classification and detection in the deep learning era. Our review encompasses major aspects underlying typical frameworks, which include concrete defect taxonomy, public datasets and evaluation metrics. We provide also a taxonomy of deep-learning-based classification and detection algorithms with a detailed discussion of their advantages and limitations. We also benchmark baseline models for classification and detection, using two popular datasets. We finally discuss important challenges of concrete defect classification and detection, and promising research avenues to build better models and integrate them in real-world visual inspection systems, which warrant further scientific investigation. IEEE},
note = {Publisher: Institute of Electrical and Electronics Engineers Inc.},
keywords = {Annotation, Annotations, Bridges, Classification, Concrete, Concrete bridge defect, Concrete bridge defects, Concrete bridges, Concrete defects, Concretes, Deep learning, Defect classification, Defect detection, Defects, Detection, Inspection, Reviews, Segmentation, Taxonomies, Visualization},
pubstate = {published},
tppubtype = {article}
}
Visual inspection is an important process for maintaining bridges in road transportation systems, and preventing catastrophic events and tragedies. In this process, accurate and automatic concrete defect classification and detection are major components to ensure early identification of any issue that can compromise the bridge safety and integrity. While a tremendous body of research has been proposed in the last decades for addressing these problems, the advent of deep learning unleashed huge opportunities for building more accurate and efficient methods. Our aim in this survey is to study the recent progress of vision-based concrete bridge defect classification and detection in the deep learning era. Our review encompasses major aspects underlying typical frameworks, which include concrete defect taxonomy, public datasets and evaluation metrics. We provide also a taxonomy of deep-learning-based classification and detection algorithms with a detailed discussion of their advantages and limitations. We also benchmark baseline models for classification and detection, using two popular datasets. We finally discuss important challenges of concrete defect classification and detection, and promising research avenues to build better models and integrate them in real-world visual inspection systems, which warrant further scientific investigation. IEEE
2.
El-Kass, W.; Gagnon, S.; Iglewski, M.
A visual and results-driven rules composition approach for better information extraction Article d'actes
Dans: A., Zaremba M. Sasiadek J. Dolgui (Ed.): IFAC-PapersOnLine, p. 112–117, 2015, ISBN: 24058963 (ISSN), (Issue: 3 Journal Abbreviation: IFAC-PapersOnLine).
Résumé | Liens | BibTeX | Étiquettes: Automation, F-score, Flow visualization, Harmonic mean, Information analysis, Information extraction, Information extraction rules, Information retrieval, Rule based, Rule composition, Rules composition, Visual process, Visualization
@inproceedings{el-kass_visual_2015,
title = {A visual and results-driven rules composition approach for better information extraction},
author = {W. El-Kass and S. Gagnon and M. Iglewski},
editor = {Zaremba M. Sasiadek J. Dolgui A.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84953876235&doi=10.1016%2fj.ifacol.2015.06.067&partnerID=40&md5=3cd38b1d6b9efc819cd882d181cdda92},
doi = {10.1016/j.ifacol.2015.06.067},
isbn = {24058963 (ISSN)},
year = {2015},
date = {2015-01-01},
booktitle = {IFAC-PapersOnLine},
volume = {28},
pages = {112–117},
abstract = {We present a highly visual process for creating and combining elementary information extraction rules, based on their results, in order to find the rules combination that produces the most accurate information extraction results. A rule's accuracy is determined by its F-Score which is the harmonic mean of the precision and the recall of that rule. Rules are combined using logical OR and AND operators. Running a few hundreds rules combinations over a corpus, in order to determine their accuracies, can take days. Using our approach, millions of rules combinations can be tested and their accuracies (F-Score) can be calculated in few seconds. A prototype was created to demonstrate the effectiveness of our approach. © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.},
note = {Issue: 3
Journal Abbreviation: IFAC-PapersOnLine},
keywords = {Automation, F-score, Flow visualization, Harmonic mean, Information analysis, Information extraction, Information extraction rules, Information retrieval, Rule based, Rule composition, Rules composition, Visual process, Visualization},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a highly visual process for creating and combining elementary information extraction rules, based on their results, in order to find the rules combination that produces the most accurate information extraction results. A rule's accuracy is determined by its F-Score which is the harmonic mean of the precision and the recall of that rule. Rules are combined using logical OR and AND operators. Running a few hundreds rules combinations over a corpus, in order to determine their accuracies, can take days. Using our approach, millions of rules combinations can be tested and their accuracies (F-Score) can be calculated in few seconds. A prototype was created to demonstrate the effectiveness of our approach. © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
3.
Allili, M.; Li, D.; Allili, M. S.
A digital topology-based method for the topological filtering of a reconstructed surface Article d'actes
Dans: Proceedings of SPIE - The International Society for Optical Engineering, San Francisco, CA, 2011, ISBN: 0277786X (ISSN); 978-081948405-5 (ISBN), (Journal Abbreviation: Proc SPIE Int Soc Opt Eng).
Résumé | Liens | BibTeX | Étiquettes: Approximation algorithms, Closing holes, Data handling, Data visualization, Digital topology, Piecewise linear techniques, Surface reconstruction, Three dimensional, Topological filtering, Topology, Visualization, Volumetric representation, Voronoi, Voronoi filtering
@inproceedings{allili_digital_2011,
title = {A digital topology-based method for the topological filtering of a reconstructed surface},
author = {M. Allili and D. Li and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-79951882558&doi=10.1117%2f12.872653&partnerID=40&md5=678176a90e6b1b3e76292d3dadaadd72},
doi = {10.1117/12.872653},
isbn = {0277786X (ISSN); 978-081948405-5 (ISBN)},
year = {2011},
date = {2011-01-01},
booktitle = {Proceedings of SPIE - The International Society for Optical Engineering},
volume = {7868},
address = {San Francisco, CA},
abstract = {In this paper, we use concepts from digital topology for the topological filtering of reconstructed surfaces. Given a finite set S of sample points in 3D space, we use the voronoi-based algorithm of Amenta & Bern 1 to reconstruct a piecewise-linear approximation surface in the form of a triangular mesh with vertex set equal to S. A typical surface obtained by means of this algorithm often contains small holes that can be considered as noise. We propose a method to remove the unwanted holes that works as follows. We first embed the triangulated surface in a volumetric representation. Then, we use the 3Dhole closing algorithm of Aktouf et al.2 to filter the holes by their size and close the small holes that are in general irrelevant to the surface while the larger holes often represent topological features of the surface. We present some experimental results that show that this method allows to automatically and effectively search and suppress unwanted holes in a 3D surface. © 2011 SPIE-IS&T.},
note = {Journal Abbreviation: Proc SPIE Int Soc Opt Eng},
keywords = {Approximation algorithms, Closing holes, Data handling, Data visualization, Digital topology, Piecewise linear techniques, Surface reconstruction, Three dimensional, Topological filtering, Topology, Visualization, Volumetric representation, Voronoi, Voronoi filtering},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, we use concepts from digital topology for the topological filtering of reconstructed surfaces. Given a finite set S of sample points in 3D space, we use the voronoi-based algorithm of Amenta & Bern 1 to reconstruct a piecewise-linear approximation surface in the form of a triangular mesh with vertex set equal to S. A typical surface obtained by means of this algorithm often contains small holes that can be considered as noise. We propose a method to remove the unwanted holes that works as follows. We first embed the triangulated surface in a volumetric representation. Then, we use the 3Dhole closing algorithm of Aktouf et al.2 to filter the holes by their size and close the small holes that are in general irrelevant to the surface while the larger holes often represent topological features of the surface. We present some experimental results that show that this method allows to automatically and effectively search and suppress unwanted holes in a 3D surface. © 2011 SPIE-IS&T.