计算机工程与应用2017，Vol.53Issue(22)：8-15,8.DOI:10.3778/j.issn.1002-8331.1708-0195
基于深度卷积神经网络的道路场景理解
Road scene understanding based on deep convolutional neural net- work
摘要
Abstract
In the self-driving technology, the road scene understanding is a very important task for environment percep-tion, and it is a challenging topic. In this paper, a deep Road Scene Segmentation Network(RSSNet)is presented, which is a 32-layer full convolutional network composed of convolution encoded network and deconvolution decoded network. The batch normalization layer used in the RSSNet prevents the vanishing gradient problem from appearing during the training process;the activation layer using the Maxout function further weakens the vanishing gradient and avoids the net-work falling into a saturated mode and neuron death phenomenon; moreover, the RSSNet using dropout operation pre-vents the over-fitting phenomenon of the network model;the max-pool indices of the feature map saved by the encoded-network are used in the decoded-network to upsample the feature map, which keeps the important edge information down. The experimental results show that the RSSNet can greatly improve the training efficiency and the segmentation accuracy, effectively classify each pixel in the road scene image and smoothly segment the objects, and provide useful information of road environment for driverless cars.关键词
深度学习/卷积神经网络/场景理解/语义分割Key words
deep learning/convolutional neural network/scenes understanding/semantic segmentation分类
信息技术与安全科学引用本文复制引用
吴宗胜,傅卫平,韩改宁..基于深度卷积神经网络的道路场景理解[J].计算机工程与应用,2017,53(22):8-15,8.基金项目
国家自然科学基金(No.10872160). （No.10872160）
