Traffic sign recognition using convolutional neural networks
Abstract
Traffic sign recognition is an important method that improves the safety in the roads, and this system is an additional step to autonomous driving. Nowadays, to solve traffic sign recognition problem, convolutional neural networks (CNN) can be adopted for its high performance well proved for computer vision applications. This paper proposes histogram equalization preprocessing (HOG) and CNN with additional operations – batch normalization, dropout and data augmentation. Several CNN architectures are compared to differentiate how each operation affects the accuracy of CNN model. Experimental results describe the effectiveness of using CNN with proposed operations.
Article in English.
Kelio ženklų atpažinimas naudojant neuroninį tinklą
Santrauka
Kelio ženklų atpažinimas – vienas iš svarbių būdų pagerinti saugumą keliuose. Ši sistema laikoma papildomu autonominio vairavimo žingsniu. Šiandien kelio ženklų atpažinimo problemai spręsti taikomi konvoliuciniai neuroniniai tinklai (KNN) dėl jų našumo, įrodyto vaizdų atpažinimo programose. Šiame straipsnyje siūlomas vaizdų histogramos išlyginimo apdorojimo metodas ir KNN su papildomomis operacijomis – paketo normalizavimas ir neuronų išjungimas / įjungimas. Yra palyginamos kelios KNN architektūros siekiant ištirti, kokią įtaką kiekviena operacija daro KNN modelio tikslumui. Eksperimentiniai rezultatai apibūdina KNN naudojimo efektyvumą su pasiūlytomis operacijomis.
Reikšminiai žodžiai: kelio ženklų atpažinimas, vaizdų apdorojimas, klasifikavimas, konvoliucinis neuroninis tinklas, paketo normalizavimas, neuronų išjungimas / įjungimas, eksperimentai.
Keyword : traffic sign recognition, image pre-processing, classification, convolutional neural network, batch normalization, dropout, experiment
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
Boujemaa, K. S., Bouhoute, A., Boubouh, K., & Berrada, I. (2017). Traffic sign recognition using convolutional neural networks. International Conference on Wireless Networks and Mobile Communications (WINCOM) (pp. 1-6). Rabat.
Budhiraja, A. (2016). Dropout in (Deep) Machine learning. Retrieved from https://medium.com/@amarbudhiraja/https-medium-com-amarbudhiraja-learning-less-to-learn-better-dropout-in-deep-machine-learning-74334da4bfc5
Chilamkurthy, S. (2017). Keras Tutorial – Traffic Sign Recognition. Retrieved from https://chsasank.github.io/keras-tutorial.html
Chollet, F. (2015). Keras. Retrieved from https://github.com/keras-team/
Ciresan, D., Meier, U., Masci, J., & Schmidhuber, J. (2011). A committee of neural networks for traffic sign classification. Proceedings of the International Joint Conference on Neural Networks. 1918-1921. 10.1109/IJCNN.2011.6033458. Retrieved from http://www.people.usi.ch/mascij/data/papers/2011_ijcnn_committee.pdf
Doukali, F. (2017). Batch normalization in Neural Networks. Retrieved from https://towardsdatascience.com/batch-normalization-in-neural-networks-1ac91516821c
Haloi, M. (2015). Traffic Sign Classification Using Deep Inception Based Convolutional Networks. CoRR, abs/1511.02992. Retrieved from https://arxiv.org/pdf/1511.02992.pdf
Hinton, G. E., Srivastava, N., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2012). Improving neural networks by preventing co-adaptation of feature detectors. CoRR abs/1207.0580.
Yadav, V. (2016). German sign classification using deep learning neural networks. Retrieved from https://chatbotslife.com/german-sign-classification-using-deep-learning-neural-networks-98-8-solution-d05656bf51ad
Yang, Y., Liu, S., Ma, W., Wang, Q., Zheng, & Liu. (2018). Efficent Traffic-Sign Recognition with Scale-aware CNN. BMVC. Retrieved from BMVC.
Yin, S., Deng, J., Zhang, D., & Du, J. (2017). Traffic Sign Recognition Based on Deep Convolutional Neural Network. CCCV, 685-695.
Lecun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989). Backpropagation applied to handwritten zip code recognition. Neural Computations, 541-551.
Loffe, S., & Szegedy, C. (2015). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. ICML.
Mao, X., Hijazi, S. L., Casas, R. A., Kaul, P., Kumar, R., & Rowen, C. (2016). Hierarchical CNN for traffic sign recognition. Intelligent Vehicles Symposium, 130-135.
Melekhov, I., Kannala, J., & Rahtu, E. (2017, October 31). Image Patch Matching Using Convolutional Descriptors with Euclidean Distance. Retrieved from https://arxiv.org/pdf/1710.11359.pdf
Pandiyan, D. (2017). Traffic Sign Classifier. Retrieved from https://github.com/dhnkrn/Traffic-Sign-Classifier
Rouse, M. (2018). Neural network. Retrieved from https://search-enterpriseai.techtarget.com/definition/neural-network
Shustanov, A., & Yakimov, P. (2017). CNN Design for Real-Time Traffic Sign Recognition. 3rd International Conference “Information Technology and Nanotechnology”. Samara.
Siddhart, D. (2017, November 17). CNNs Architectures: LeNet, AlexNet, VGG, GoogLeNet, ResNet and more …. Retrieved from https://medium.com/@siddharthdas_32104/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-and-more-666091488df5
Srivastava, N., Hinton, G. E., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(1), 1929-1958.
Stallkamp, J., Schlipsing, M., Salmen, J., & Igel, C. (2012, February). Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognition. Retrieved from http://www.sciencedirect.com/science/article/pii/S0893608012000457
Voelcker, J. (2014). 1.2 Billion Vehicles On World’s Roads Now, 2 Billion By 2035: Report. Retrieved from https://www.greencarreports.com/news/1093560_1-2-billion-vehicles-on-worlds-roads-now-2-billion-by-2035-report
Strongtie Insurance. (2018). What Are The Most Common Reasons for Road Accidents? Retrieved from https://www.strongtieinsurance.com/common-reasons-road-accidents/