Road surface condition classification using deep learning

Abstract Traditional image recognition technology currently cannot achieve the fast real-time high-accuracy performance necessary for road recognition in intelligent driving. Deep learning models have been recently emerging as promising tools to achieve this performance. The recognition performance of such models can be boosted using appropriate selection of the activation functions. This paper proposes a deep learning approach for the classification of road surface conditions, and constructs a new activation function based on the rectified linear unit Rectified Linear Units (ReLu) activation function. The experimental results show a classification accuracy of 94.89% on the road state database. Experiments on public datasets demonstrate that the proposed convolutional neural network model with the improved activation function has better generalization and excellent classification performance.

[1]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[2]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[3]  Liu Hongwei,et al.  Target Detection Method Based on Convolutional Neural Network for SAR Image , 2016 .

[4]  Yungang Liu,et al.  The application of deep learning in computer vision , 2017, 2017 Chinese Automation Congress (CAC).

[5]  James M. Keller,et al.  Multilayer Neural Networks and Backpropagation , 2016 .

[6]  Shunli Zhang,et al.  Single Object Tracking With Fuzzy Least Squares Support Vector Machine , 2015, IEEE Transactions on Image Processing.

[7]  Chris Aldrich,et al.  Flotation froth image recognition with convolutional neural networks , 2019, Minerals Engineering.

[8]  Paul Newman,et al.  1 year, 1000 km: The Oxford RobotCar dataset , 2017, Int. J. Robotics Res..

[9]  Yee Whye Teh,et al.  Unsupervised Discovery of Nonlinear Structure Using Contrastive Backpropagation , 2006, Cogn. Sci..

[10]  W Burghout,et al.  IMAGE-BASED WINTER ROAD CONDITION RECOGNITION , 1998 .

[11]  Hari Krishna Vydana,et al.  Investigative study of various activation functions for speech recognition , 2017, 2017 Twenty-third National Conference on Communications (NCC).

[12]  D. Hubel,et al.  EFFECTS OF VISUAL DEPRIVATION ON MORPHOLOGY AND PHYSIOLOGY OF CELLS IN THE CATS LATERAL GENICULATE BODY. , 1963, Journal of neurophysiology.

[13]  Yi-Chao Wu,et al.  Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models , 2017, Pattern Recognit..

[14]  Andreas Geiger,et al.  Vision meets robotics: The KITTI dataset , 2013, Int. J. Robotics Res..

[15]  D. Harvey.  RELU Special Issue: Editorial Reflections , 2006 .

[16]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[17]  Yang Yu,et al.  Color Images Illumination Compensation Method for Human Face Region Detection , 2014 .

[18]  Philipp Petersen,et al.  Optimal approximation of piecewise smooth functions using deep ReLU neural networks , 2017, Neural Networks.

[19]  Chang Luo,et al.  A novel softplus linear unit for deep convolutional neural networks , 2017, Applied Intelligence.

[20]  Chengjun Liu,et al.  Extracting Multiple Features in the CID Color Space for Face Recognition , 2010, IEEE Transactions on Image Processing.

[21]  Hong Bao,et al.  Forward Vehicle Detection Based on Incremental Learning and Fast R-CNN , 2017, 2017 13th International Conference on Computational Intelligence and Security (CIS).