Research on Real-Time Semantic Segmentation Algorithm for Driverless Scene
Aiming at the problem that it is difficult to achieve a balance between segmentation accuracy and run-ning speed in existing unmanned driving scenarios,a new real-time semantic segmentation algorithm is proposed.First,an Efficient Feature Extraction Unit(EFEU)is constructed using the residual structure(Residual)to better per-ceive spatial and semantic information.Secondly,a bilateral structure is adopted,in which the spatial branch combined with the pooling operation preserves the shallow spatial information,and the context branch is used to pro-vide a large receptive field and capture deep contextual information.These two branches are merged at different stages of the network to enhance the information propagation between different layers.Finally,a context fusion module is constructed by combining depth-wise separable convolution and channel shuffle to fuse information between different levels to further improve the segmentation effect of the model.This paper conducts experiments on commonly used datasets,and the experimental results show the effectiveness of the proposed method.
NETL
NSTL
万方数据
维普
