Neural network-based semantic segmentation model for robot perception of driverless vision

作     者：Lu Ye Ting Duan Jiayi Zhu 

作者机构：School of Information and Electronic EngineeringZhejiang University of Science&TechnologyHangzhouZhejiangPeople's Republic of China School of Mechanical and Energy EngineeringZhejiang University of Science&TechnologyHangzhouZhejiangPeople's Republic of China 

出 版 物：《IET Cyber-Systems and Robotics》 (智能系统与机器人（英文）)

年 卷 期：2020年第2卷第4期

页      面：190-196页

核心收录：

学科分类：08[工学] 081101[工学-控制理论与控制工程] 0811[工学-控制科学与工程] 081102[工学-检测技术与自动化装置] 

基　　金：College of Agricultural and Life Sciences, University of Wisconsin NIMH-PHS(MH19689) Office of Economic Research Economic Development Administration(99-7-13248, OER-417-G-72-7) 

主　　题：robot computer Driver 

摘      要：Driverless vision is one of the important applications of robot *** the development of driverless vehicles,the perception and understanding of the surrounding environment are becoming more and more *** the types of surrounding objects are too complex,the ability of the computer to recognise the environment is *** improve the recognition accuracy of the computer and enhance the ability of segmentation,in this study,depth estimation is used to predict depth information to assist semantic segmentation,and then edge features of objects are introduced to enhance the contour of objects.A neural network-based semantic segmentation model is ***,the intrinsic mechanism of attention is used to increase the correlation between *** experimental results on the CamVid data set show that this model can obtain better evaluation results and improve the segmentation accuracy of images compared with other models.