基于深度学习的变电站巡检机器人道路场景识别

为了提升变电站巡检机器人对自身所处环境的理解能力,将深度学习技术应用于变电站巡检机器人对道路场景的识别中,提出了一种全卷积道路场景识别网络(road scene recognition net,RSRNet)。该网络主要由相对浅层的编码网络和镜像结构与跳层融合结构相结合的解码网络组成,通过编码网络提取图像特征后由解码网络识别出图像目标信息。通过实验表明,本文提出的网络在同类型网络中识别精度及效率更高,同时在实际变电站场景中也表现出了优良的场景识别性能。

Road Scene Recognition of Substation Inspection Robot Based on Deep Learning

In order to improve the understanding ability to the environment of substation inspection robot, a fully convolutional road scene recognition network (RSRNet) based on deep learning technology is used to recognize substation road scene. RSRNet is composed of shallow encoder network and decoder network combining of mirror structure and skip-level fusion structure. Encoder network is used to extract image features, decoder network is used to recognize image information. Experimental results show that the proposed network has higher recognition accuracy and efficiency compared to the same type of network. At the same time, it also shows excellent scene recognition performance in actual substation environment.