Evidential Data Fusion for Characterization of Pavement Surface Conditions during Winter Using a Multi-Sensor Approach
Abstract
:1. Introduction
2. State of the Art
3. Materials and Methods
3.1. Data
- 56,000 RGB camera images.
- 75 scenes of 50 to 100 images each.
- 10 classes of annotation.
- Adverse weather driving conditions, including snow.
3.2. Image Classification
3.3. Audio Classification
3.4. Multi-Sensor Fusion System
3.4.1. Modelling
3.4.2. Estimation
3.4.3. Combination
3.4.4. Decisions
3.4.5. Development
4. Results
4.1. Image Classification
- Presence of both snow and water on the road surface.
- Presence of tree shade.
- Large white surface markings were confused with snow.
- Low ambient light levels.
- Wet ground with light reflection giving the impression of ice.
- Road partially covered with snow, water, or ice.
- Confusion between white ice and snow because of the color.
4.2. Audio Classification
4.3. Multisensor Data Fusion
4.3.1. Environment #1 (Black Ice)
4.3.2. Environment #2 (Melting Snow)
5. Discussion
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Jonsson, P.; Casselgren, J.; Thörnberg, B. Road surface status classification using spectral analysis of NIR camera images. IEEE Sens. J. 2015, 15, 16. [Google Scholar] [CrossRef]
- U.S. Department of Transportation Federal Highway Administration. Available online: https://highways.dot.gov/ (accessed on 12 November 2021).
- 2007–2016 10-Year Averages Analyzed by Booz Allen Hamilton, based on NHTSA Data in U.S. Department of Transportation Federal Highway Administration. Available online: https://ops.fhwa.dot.gov/weather/q1_roadimpact.htm (accessed on 12 November 2021).
- Morin, E. Optimisation de la Gestion de L’information Météo-Routière pour le Ministère des Transports du Québec—Direction de l’Estrie; Mémoire de Maîtrise; Université de Sherbrooke: Sherbrooke, QC, Canada, 2010. [Google Scholar]
- Khoderagha, N. Outil Innovant pour la Gestion des Routes; Mémoire de Maîtrise, Université du Québec: Montréal, QC, Canada, 2019. [Google Scholar]
- Zagvozda, M.; Dimter, S.; Moser, V.; Barišić, I. Application of GIS technology in pavement management systems. J. Croat. Assoc. Civ. Eng. 2019, 71, 297–304. [Google Scholar]
- Tarleton, J. Critical conditions: The weather plays a vital role in road safety and traffic management applications. Intertraffic World Annual Showcase 2015. Traffic Manag. 2015, pp. 108–109. Available online: https://www.vaisala.com/sites/default/files/documents/WEA-RDS-G-TTI_Intertraffic%20World_2014_final.pdf (accessed on 12 November 2021).
- Teconer. Available online: https://www.teconer.fi/en/ (accessed on 12 November 2021).
- High Sierra. Available online: https://hsierra.com/product/remote-road-surface-condition-sensor-series-5433/ (accessed on 12 November 2021).
- Lufft/OTT Hydromet. Available online: https://www.otthydromet.com/en/Ott/p-lufft-starwis-umb-non-invasive-road-condition-sensor-5-m-distance/8711.U55 (accessed on 12 November 2021).
- Campbell Scientific. Available online: https://www.campbellsci.ca/ccfc (accessed on 12 November 2021).
- Chagnon, F. Caractérisation des États de Surface par Télédétection Infrarouge Thermique Multispectrale: Contribution à L’étude des Conditions de Viabilité Hivernale. Ph.D. Thesis, Université de Sherbrooke, Sherbrooke, QC, Canada, 2008; p. 130. [Google Scholar]
- Song, I.Y.; Yoon, J.H.; Bae, S.H.; Jeon, M.; Shin, V. Classification of road surface status using a 94 GHz dual-channel polarimetric radiometer. Int. J. Remote Sens. 2012, 33, 5746–5767. [Google Scholar] [CrossRef]
- Gui, K.; Ye, L.; Ge, J.; Cheikh, F.A.; Huang, L. Road surface condition detection utilizing resonance frequency and optical technologies. Sens. Actuators A Phys. 2019, 297, 10. [Google Scholar] [CrossRef]
- Tong, Z.; Xu, P.; Denœux, T. An evidential classifier based on Dempster-Shafer theory and deep learning. Neurocomputing 2021, 450, 275–293. [Google Scholar] [CrossRef]
- Tong, Z.; Xu, P.; Denœux, T. Fusion of Evidential CNN Classifiers for Image Classification. In Belief Functions: Theory and Applications. BELIEF; Denœux, T., Lefèvre, E., Liu, Z., Pichon, F., Eds.; Lecture Notes in Computer Science; Springer: Cham, Switzerland, 2021; p. 12915. [Google Scholar]
- Shafer, G. A mathematical theory of evidence, In Foundations of Probability Theory, Statistical Inference, and Statistical Theories of Science; Springer: Dordrecht, The Netherlands, 1976; pp. 365–436. [Google Scholar]
- Dempster, A.P. A generalization of Bayesian inference. J. R. Stat. Soc. Ser. B 1968, 30, 205–2047. [Google Scholar] [CrossRef]
- Ruta, A.; Li, Y.; Liu, X. Real-time traffic sign recognition from video by class-specific discriminative features. Pattern Recognit. 2010, 43, 416–430. [Google Scholar] [CrossRef]
- Yi, S.-C.; Chen, Y.-C.; Chang, C.-H. A lane detection approach based on intelligent vision. Comput. Electr. Eng. 2015, 42, 23–29. [Google Scholar] [CrossRef]
- Mataei, B.; Moghadas Nejad, F.; Zahedi, M.; Zakeri, H. Evaluation of pavement surface drainage using an automated image acquisition and processing system. Autom. Constr. 2018, 86, 240–255. [Google Scholar] [CrossRef]
- Khan, M.N.; Das, A.; Ahmed, M.M.; Wulff, S.S. Multilevel weather detection based on images: A machine learning approach with histogram of oriented gradient and local binary pattern-based features. J. Intell. Transp. Syst. 2021, 25, 513–532. [Google Scholar] [CrossRef]
- Kongrattanaprasert, W.; Nomura, H.; Kamakura, T.; Ueda, K. Detection of road surface conditions using tire noise from vehicles. IEEJ Trans. Ind. Appl. 2009, 129, 761–767. [Google Scholar] [CrossRef]
- Winroth, J.; Kropp, W.; Hoever, C.; Beckenbauer, T.; Maennel, M. Investigating generation mechanisms of tyre/road noise by speed exponent analysis. Appl. Acoust. 2017, 115, 101–108. [Google Scholar] [CrossRef]
- Ramos-Romero, C.; León-Ríos, P.; Al-Hadithi, B.M.; Sigcha, L.; de Arcas, G.; Asensio, C. Identification and mapping of asphalt surface deterioration by tyre-pavement interaction noise measurement. Measurement 2019, 146, 718–727. [Google Scholar] [CrossRef]
- Bochra, G.; Safia, S. Les Réseaux de Neurones Convolutionels (CNN) Pour la Classification des Images Associées Aux Places de Stationnement d’un Parc de Véhicule; Mémoire de Maîtrise, Djlali Bounaama: Khemis Miliana, Algerie, 2019. [Google Scholar]
- Simonnet, E. Réseaux de Neurones Profonds Appliqués à la Compréhension de la Parole. Ph.D. Thesis, Le Mans Université, Le Mans, France, 2019. [Google Scholar]
- Pan, G.; Fu, L.; Yu, R.; Muresan, M. Winter Road Surface Condition Recognition Using a Pre-Trained Deep Convolutional Neural Network; IEEE: Piscataway, NJ, USA, 2019; pp. 614–620. [Google Scholar]
- Abdic, I.; Fridman, L.; Brown, D.E.; Angell, W.; Reimer, B.; Marchi, E.; Schuller, B. Detecting road surface wetness from audio: A deep learning approach. In Proceedings of the 23rd International Conference on Pattern Recognition (ICPR 2016), Cancun, Mexico, 5–8 December 2016; pp. 3458–3463. [Google Scholar]
- White, F.E. Data Fusion Lexicon: Fort Belvoir; Defense Technical Information Center: Fort Belvoir, VA, USA, 1991; p. 16. [Google Scholar]
- Bloch, I.; Maitre, H. École Nationale Supérieure des Télécommunications-CNRS UMR. Les Méthodes de Raisonnement dans les Images. 2004. Available online: https://perso.telecom-paristech.fr/bloch/VOIR/poly_voir.pdf (accessed on 10 February 2020).
- André, C.; Hégarat-Mascle, S.L.; Reynaud, R. Evidential framework for data fusion in a multi-sensor surveillance system. Eng. Appl. Artif. Intell. 2015, 43, 166–180. [Google Scholar] [CrossRef]
- Bezerra, E.D.C.; Teles, A.S.; Coutinho, L.R.; da Silva e Silva, F.J. Dempster-Shafer Theory for modeling and treating uncertainty in IoT applications based on complex event processing. Sensors 2021, 21, 1863. [Google Scholar] [CrossRef] [PubMed]
- Irschik, D.; Stork, W. Road surface classification for extended floating car data. In Proceedings of the International Conference on Vehicular Electronics and Safety, Hyderabad, India, 16–17 December 2014; pp. 78–83. [Google Scholar]
- Zhao, J.; Wu, H.; Chen, L. Road surface state recognition based on SVM optimization and image segmentation processing. J. Adv. Transp. 2017, 2017, 1–21. [Google Scholar] [CrossRef]
- Pitropov, M.; Garcia, D.; Rebello, J.; Smart, M.; Wang, C.; Czarnecki, K.; Waslander, S. Canadian Adverse Driving Conditions Dataset. Int. J. Robot. Res. 2021, 40, 681–690. [Google Scholar] [CrossRef]
- Center for Data Innovation. Available online: https://datainnovation.org/2018/08/detecting-weather-conditions-on-the-road/ (accessed on 11 May 2021).
- Association des Piétons et Cyclistes Pont Jacques-Cartier. Available online: flickr.com/photos/pontjacquescartier (accessed on 20 January 2020).
- Kongrattanaprasert, W.; Nomura, H.; Kamakura, T.; Ueda, K. Detection of road surface states from tire noise using Neural Network Analysis. IEEJ Trans. Ind. Appl. 2010, 130, 920–925. [Google Scholar] [CrossRef]
- Li, T. A state-of-the-art review of measurement techniques on tire–pavement interaction noise. Measurement 2018, 28, 325–351. [Google Scholar] [CrossRef]
- Jay, P. Understanding and Implementing Architectures of ResNet and ResNeXt for State-of-the-Art Image Classification: From Microsoft to Facebook, Part 1. Available online: https://medium.com/@14prakash/understanding-and-implementing-architectures-of-resnet-and-resnext-for-state-of-the-art-image-cf51669e1624 (accessed on 11 May 2021).
- Iandola, F.N.; Han, S.; Moskewicz, M.W.; Ashraf, K.; Dally, W.J.; Keutzer, K. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size. arXiv 2016, arXiv:1602.07360. [Google Scholar]
- Howard, A.; Sandler, M.; Chu, G.; Chen, L.C.; Chen, B.; Tan, M.; Wang, W.; Zhu, Y.; Pang, R.; Vasudevan, V.; et al. Searching for mobilenetv3. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Korea, 27 October–2 November 2019; pp. 1314–1324. [Google Scholar]
- Dai, W.; Dai, C.; Qu, S.; Li, J.; Das, S. Very deep convolutional neural networks for raw waveforms. In Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA, 5–9 March 2017; pp. 421–425. [Google Scholar]
- Herring, W. Audio Classifier Tutorial. Available online: https://colab.research.google.com/github/pytorch/tutorials/blob/gh-pages/_downloads/audio_classifier_tutorial.ipynb (accessed on 22 March 2021).
- Abbas, N. Développement de Modèles de Fusion et de Classification Contextuelle D’images Satellitaires par la Théorie de L’évidence et la Théorie du Raisonnement Plausible et Paradoxal. Ph.D. Thesis, Université Des Sciences Et De La Technologie Houari Boumediene, Bab Ezzouar, Algérie, 2009; p. 76. [Google Scholar]
- Germain, M. Classification Multisource par la Fusion Évidentielle avec une Nouvelle Approche Statistique Floue. Ph.D. Thesis, Université de Sherbrooke, Sherbrooke, QC, Canada, 2006; p. 173. [Google Scholar]
- An Introduction to DSmT; Applications and Advances of DSmT for Information Fusion; Smarandache, F.; Dezert, J. (Eds.) American Research Press: Rehoboth, AL, USA, 2009; Volume 3, p. 45. [Google Scholar]
- Dong, Y.; Li, X.; Dezert, J.; Khyam, M.O.; Noor-A-Rahim, M.; Ge, S.S. DSmT-based fusion strategy for human activity recognition in body sensor networks. IEEE Trans. Ind. Inform. 2020, 16, 7138–7149. [Google Scholar] [CrossRef]
- Walia, G.S.; Gupta, A.; Kapoor, R. Intelligent fire-detection model using statistical color models data fusion with Dezert–Smarandache method. Int. J. Image Data Fusion 2013, 4, 324–341. [Google Scholar] [CrossRef]
- Smets, P. Data Fusion in the Transferable Belief Model. In Proceedings of the Third International Conference on Information Fusion, Paris, France, 10–13 July 2000; pp. 21–33. [Google Scholar]
Model | Number of Layers | Number of Parameters |
---|---|---|
ResNet50 | 50 | 25.6 M |
MobileNetV3_small | 16 | 1.522 M |
SqueezeNet1_1 | 11 | 724,548 k |
Architecture | Epoch | Training Accuracy | Validation Accuracy |
---|---|---|---|
ResNet50 | 250 | 99.9% | 99.1% |
MobilenetV3_small | 100 | 99.5% | 99.0% |
Squeezenet1_1 | 80 | 99.7% | 98.8% |
Video Architecture | Duration (s) | Images per Second |
---|---|---|
Resnet50 | 347 | 1.9 |
Mobilenet | 237 | 2.8 |
Squeezenet | 170 | 3.9 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Diaby, I.; Germain, M.; Goïta, K. Evidential Data Fusion for Characterization of Pavement Surface Conditions during Winter Using a Multi-Sensor Approach. Sensors 2021, 21, 8218. https://doi.org/10.3390/s21248218
Diaby I, Germain M, Goïta K. Evidential Data Fusion for Characterization of Pavement Surface Conditions during Winter Using a Multi-Sensor Approach. Sensors. 2021; 21(24):8218. https://doi.org/10.3390/s21248218
Chicago/Turabian StyleDiaby, Issiaka, Mickaël Germain, and Kalifa Goïta. 2021. "Evidential Data Fusion for Characterization of Pavement Surface Conditions during Winter Using a Multi-Sensor Approach" Sensors 21, no. 24: 8218. https://doi.org/10.3390/s21248218
APA StyleDiaby, I., Germain, M., & Goïta, K. (2021). Evidential Data Fusion for Characterization of Pavement Surface Conditions during Winter Using a Multi-Sensor Approach. Sensors, 21(24), 8218. https://doi.org/10.3390/s21248218