Abstract
Automated driving gradually emerges as a real reality, but it still has to face various challenges, including sophisticated and volatile traffic conditions, human operating faults, etc. Amongst them, accurate understanding of traffic signs by using computer vision and deep learning methods has great significance for driving safety. In recent years, the advent of deep learning has made this issue much effectively. In this article, our major goal is to use deep learning models for conducting traffic-light sign recognition related to autonomous vehicles. As far as we know, this is the first time that the capsule neural network is employed as a method for scene understanding so as to effectively identify a class of traffic-light signs. Compared with the well-known convolutional neural networks, capsule networks diminish the demands for training datasets and tackle the spatial relationship much precisely.
Similar content being viewed by others
References
Bach M, Stumper D, Dietmayer K (2018) Deep convolutional traffic light recognition for automated driving. Intelligent Transportation Systems, 851–858
Behrendt K, Novak L, Botros R (2017) A deep learning approach to traffic lights: Detection, tracking, and classification. IEEE International Conference on Robotics and Automation (ICRA), Singapore, 1370–1377
Bernstein D (2010) Essentials of psychology. Cengage Learning, 123–124
Brefczynski-Lewis JA, Lewis JW (2017) Auditory object perception: A neurobiological model and prospective review. Neuropsychologia, 223–242
Han C, Gao G, Zhang Y (2018) Real-time small traffic sign detection with revised faster-RCNN. Springer Multimedia Tools and Applications. 78
Husain F, Dellen B, Torras C (2017) Scene Understanding Using Deep Learning. Academic Press, 373–382
Jeon HS, Kum DS, Jeong WY (2018) Traffic scene prediction via deep learning: Introduction of multichannel occupancy grid map as a scene representation. IEEE Intelligent Vehicles Symposium (IV), 1496–1501
John V, Yoneda K, Qi B, Liu Z, Mita S (2014) Traffic light recognition in varying illumination using deep learning and saliency map. Intelligent Transportation Systems, 2286–2291
Kheradpisheh S, Ghodrati M, Ganjtabesh M et al (2016) Deep networks can resemble human feed-forward vision in invariant object recognition. Nature Sci Report 6:32672
Kim M, Chi S (2019) Detection of centerline crossing in abnormal driving using CapsNet. Journal of Supercomputer, 189–196
Kim H, Park J, Jung H (2018) An efficient color space for deep learning based traffic light recognition. Advanced Transportation, pp.1–12
Kim Y, Wang P, Zhu Y, Mihaylova L (2019) A capsule network for traffic speed prediction in complex road networks. Sensor Data Fusion: Trends, Solutions, Applications (SDF), 1–6
Kwabena M, Felix A, Abra A, Edward Y (2019) Capsule networks – A survey. Journal of King Saud University - Computer and Information Sciences, 1–16
Lewicki MS, Olshausen BA, Surlykke A, Moss CF (2014) Scene analysis in the natural environment. Frontiers in Psychology, 199
Malcolm GL, Groen I, Baker CI (2016) Making sense of real-world scenes. Trends in Cognitive Sciences, 843–856
Morris T (2014) Computer vision and image processing. Palgrave Macmillan
Mukhometzianov R, Carrillo J (2018) CapsNet comparative performance evaluation for image classification. Computational Intelligence and Security, 1–14
Müller J, Dietmayer K (2018) Detecting traffic lights by single shot detection. Intelligent Transportation Systems, 266–273
Nandi D, Saif AS, Prottoy P, Zubair KM, Shubho SA (2018) Traffic sign detection based on color segmentation of obscure image candidates: a comprehensive study. Int J Mod Educ Comput Sci 10(6):35–46
Park H, Jang S, Jeong H, Ha Y (2019) Roadway image preprocessing for deep learning-based driving scene understanding. IEEE International Conference on Big Data and Smart Computing (BigComp), pp. 1–4
Peixinho AZ, Benato BC, Nonato LG, Falcão AX (2018) Delaunay triangulation data augmentation guided by visual analytics for deep learning. SIBGRAPI Conference on Graphics, Patterns and Images, 384–391
Qiao K, Chen J, Wang L, Zhang C, Zeng L, Tong L, Yan B (2019) Category decoding of visual stimuli from human brain activity using a bidirectional recurrent neural network to simulate bidirectional information flows in human visual cortices. Frontiers in Neuroscience, 692
Qu H, Zhang L, Wu X, He X, Hu X, Wen X (2019) Multiscale object detection in infrared streetscape images based on deep learning and instance level data augmentation. Applied Sciences, 553–565
Saadna Y, Behloul A (2017) An overview of traffic sign detection and classification methods. Multimedia Information Retrieval, 1–18
Sabour S, Frosst N, Hinton GE (2017) Dynamic routing between capsules. International Conference on Neural Information Processing Systems, pp.3859–386
Sermanet P, LeCun Y (2011) Traffic sign recognition with multi-scale convolutional networks. International Joint Conference, 2809–2813
Stivaktakis R, Tsagkatakis G, Tsakalides P (2019) Deep learning for multilabel land cover scene categorization using data augmentation. IEEE Geosci Remote Sens Lett 16(7):1031–1035
Tampubolon H, Yang C, Chan S, Sutrisno H, Hua K-L (2019) Optimized CapsNet for traffic jam speed prediction using mobile sensor data under urban swarming transportation. Sensors, 5277
Tran T, Pham C, Phuoc N, Duong T, Jeon J (2016) Real-time traffic light detection using color density. IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia) 1–4
Tsoi T, Wheelus C (2020) Traffic signal classification with cost-sensitive deep learning models. IEEE International Conference on Knowledge Graph (ICKG), 586–592
Wali SB, Abdullah MA, Hannan MA, Hussain A, Samad SA, Ker PJ, Mansor MB (2019) Vision-based traffic sign detection and recognition systems: Current trends and challenges. Sensors (Basel), 2093
Wu N, Fang H (2017) A novel traffic light recognition method for traffic monitoring systems. Asia-Pacific Conference on Intelligent Robot Systems, 141–145
Zhang F, Wang Y, Ye M (2018) Network traffic classification method based on improved capsule neural network. International Conference on Computational Intelligence and Security, 174–178
Zhang Z, Zhang D, Wei H (2019) Vehicle type recognition using capsule network. Chinese Control and Decision Conference, 2944–2948
Zhang Y, Li J, Guo Y, Xu C, Bao C, Song Y (2019) Vehicle driving behavior recognition based on multi-view convolutional neural network with joint data augmentation. IEEE Trans Veh Technol 68(5):4223–4234
Zhao Z, Zheng P, Xu S, Wu X (2019) Object detection with deep learning: A review. Neural Networks and Learning Systems, 3212–3232
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, X., Yan, W.Q. Traffic-light sign recognition using capsule network. Multimed Tools Appl 80, 15161–15171 (2021). https://doi.org/10.1007/s11042-020-10455-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-020-10455-x