Traffic signs Detection and Recognition based on deep learning using embedded systems
Subject Areas : information technology
peyman
Babaei
1
(Ph.D., Assistant Professor, Islamic Azad University, Tehran West Branch, Computer Department, Tehran, Iran)
Faezeh
Allameh
2
(Islamic Azad University)
Keywords: Traffic signs Detection and Recognition, Embedded systems, Convolutional neural networks.,
Abstract :
Detection and recognition of traffic signs requires the use of classification algorithms, and they basically use visual information such as the shape and color of traffic signs. However, these algorithms face problems in real-time tests, and it is very difficult to achieve the detection of multiple targets, and it is necessary to accelerate the performance of the corresponding algorithms. Traffic sign recognition systems based on deep neural networks may have limitations in practical applications due to computational requirements and resource consumption. Most embedded systems interact directly with processes or the environment and make decisions based on their inputs. This makes the system reactive and responds in real-time to processing inputs to ensure proper operation. This paper presents a lightweight neural network for traffic sign recognition that achieves high accuracy and precision with fewer trainable parameters. To evaluate the efficiency and reliability of the proposed model based on convolutional neural network for traffic sign recognition, extensive experiments have been conducted on the GTSRB dataset. Also, the obtained results have been compared with several advanced architectures such as VGG16, MobileNetv2 and ResNet50. The results show that the proposed model has achieved good performance and emphasizes its potential for deploying real-time traffic sign recognition models and driving assistance systems. The computational efficiency and small size of the proposed model make it more practical and suitable for real-time traffic sign recognition.
L. Jiao et al., “A Survey of Deep Learning-Based Object Detection,” IEEE Access, vol. 7, pp. 128837–128868, 2019.
Li, J.; Wang, Z. Real-time traffic sign recognition based on efficient CNNs in the wild. IEEE Trans. Intell. Transp. Syst. 2018, 20, 975–984.
Zhu, Y.; Yan, W.Q. Traffic sign recognition based on deep learning. Multimed. Tools Appl. 2022, 81, 17779–17791.
Shustanov, A.; Yakimov, P. CNN design for real-time traffic sign recognition. Procedia Eng. 2017, 201, 718–725.
Bangquan, X.; Xiong, W.X. Real-time embedded traffic sign recognition using efficient convolutional neural network. IEEE Access 2019, 7, 53330–53346.
Zaibi, A.; Ladgham, A.; Sakly, A. A lightweight model for traffic sign classification based on enhanced LeNet-5 network. J. Sensors 2021, 2021, 8870529.
Sokipriala, J.; Orike, S. Traffic sign classification comparison between various convolution neural network models. Int. J. Sci. Eng. Res. 2021, 12, 165–171.
Mishra, J.; Goyal, S. An effective automatic traffic sign classification and recognition deep convolutional networks. Multimed. Tools Appl. 2022, 81, 18915–18934.
Chen, L.; Zhao, G.; Zhou, J.; Kuang, L. Real-time traffic sign classification using combined convolutional neural networks. In Proceedings of the 2017 4th IAPR Asian Conference on Pattern Recognition (ACPR), Nanjing, China, 26–29 November 2017; pp. 399–404.
Haque, W.A.; Arefin, S.; Shihavuddin, A.; Hasan, M.A. DeepThin: A novel lightweight CNN architecture for traffic sign recognition without GPU requirements. Expert Syst. Appl. 2021, 168, 114481.
Fang, H.F.; Cao, J.; Li, Z.Y. A small network MicronNet-BF of traffic sign classification. Comput. Intell. Neurosci. 2022, 2022, 3995209.
Sarku, E.; Steele, J.; Ruffin, T.; Gokaraju, B.; Karimodini, A. Reducing Data Costs-Transfer Learning Based Traffic Sign Classification Approach. In Proceedings of the SoutheastCon 2021, Atlanta, GA, USA, 10–13 March 2021; pp. 1–5.
Cao, J.; Song, C.; Peng, S.; Xiao, F.; Song, S. Improved traffic sign detection and recognition algorithm for intelligent vehicles. Sensors 2019, 19, 4021.
Fu, H.; Wang, H. Traffic Sign Classification Based on Prototypes. In Proceedings of the 2021 16th International Conference on Intelligent Systems and Knowledge Engineering (ISKE), Chengdu, China, 26–28 November 2021; pp. 7–10.
Sichkar, V.; Kolyubin, S. Effect of various dimension convolutional layer filters on traffic sign classification accuracy. Sci. Tech. J. Inf. Technol. Mech. Opt. 2019, 19, 546–552.
Agarwal, S.; X, C.; Kumar, R. Convolutional Neural Network for Traffic Sign Classification. Int. J. Inf. Technol. Proj. Manag. 2022, 9.
Youssouf, N. Traffic sign classification using CNN and detection using faster-RCNN and YOLOV4. Heliyon 2022, 8.
Gökberk, A.; Durdu, A.; Nesimio ˘glu, B.S. Accuracy Comparison of CNN Networks on GTSRB Dataset. J. Artif. Intell. Data Sci. 2022, 2, 63–68.
Kuros, S.; Kryjak, T. Traffic Sign Classification Using Deep and Quantum Neural Networks. 2022.
Pradana, A.I.; Rustad, S.; Shidik, G.F.; Santoso, H.A. Indonesian Traffic Signs Recognition Using Convolutional Neural Network. In Proceedings of the 2022 International Seminar on Application for Technology of Information and Communication (iSemantic), Semarang, Indonesia, 17–18 September 2022; pp. 426–430.
Bhatt, N.; Laldas, P.; Lobo, V.B. A Real-Time Traffic Sign Detection and Recognition System on Hybrid Dataset using CNN. In Proceedings of the 2022 7th International Conference on Communication and Electronics Systems (ICCES), India, June 2022; pp. 1354–1358.
