Complex Traffic Scene Image Classification Based on Sparse Optimization Boundary Semantics Deep Learning

作     者：ZHOU Xiwei WANG Huifeng LI Saisai PENG Haonan WU Jianfeng ZHOU Xiwei;WANG Huifeng;Li Saisai;PENG Haonan;WU Jianfeng

作者机构：School of Electronic&Control EngineeringChang'an UniversityXi'an 710064ShaanxiChina 

出 版 物：《Wuhan University Journal of Natural Sciences》 (武汉大学学报（自然科学英文版）)

年 卷 期：2023年第28卷第2期

页      面：150-162页

核心收录：

学科分类：1305[艺术学-设计学（可授艺术学、工学学位）] 13[艺术学] 081104[工学-模式识别与智能系统] 08[工学] 0804[工学-仪器科学与技术] 0838[工学-公安技术] 081101[工学-控制理论与控制工程] 0811[工学-控制科学与工程] 

基　　金：Supported in part by the Shaanxi Natural Science Basic Research Program(2022JM-298) the National Natural Science Foundation of China(52172324) Shaanxi Provincial Key Research and Development Program(2021SF-483) the Science and Technology Project of Shaan Provincal Transportation Department(21-202K,20-38T) 

主　　题：traffic scene SegNet image classification simple linear iterative clustering(SLIC) conditional random field boundary number 

摘      要：With the rapid development of intelligent traffic information monitoring technology,accurate identification of vehicles,pedestrians and other objects on the road has become particularly ***,in order to improve the recognition and classification accuracy of image objects in complex traffic scenes,this paper proposes a segmentation method of semantic redefine segmentation using image boundary ***,we use the Seg Net semantic segmentation model to obtain the rough classification features of the vehicle road object,then use the simple linear iterative clustering(SLIC)algorithm to obtain the over segmented area of the image,which can determine the classification of each pixel in each super pixel area,and then optimize the target segmentation of the boundary and small areas in the vehicle road ***,the edge recovery ability of condition random field(CRF)is used to refine the image *** experimental results show that compared with FCN-8s and Seg Net,the pixel accuracy of the proposed algorithm in this paper improves by 2.33%and 0.57%,*** compared with Unet,the algorithm in this paper performs better when dealing with multi-target segmentation.