深度学习背景下的图像语义分割方法综述  

Survey of image semantic segmentation methods in the deep learning era

作　　者：严毅 邓超 李琳[2] 朱凌坤 叶彪 Yan Yi;Deng Chao;Li Lin;Zhu Lingkun;Ye Biao(School of Automobile and Traffic Engineering,Wuhan University of Science and Technology,Wuhan 430063,China;School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430063,China;School of Transportation and Logistics Engineering,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉科技大学汽车与交通工程学院,武汉430063 [2]武汉科技大学计算机科学与技术学院,武汉430063 [3]武汉理工大学交通与物流工程学院,武汉430063

出　　处：《中国图象图形学报》2023年第11期3342-3362,共21页Journal of Image and Graphics

基　　金：国家自然科学基金青年基金项目(52002298);“运输车辆检测、诊断与维修技术”交通行业重点实验室开放课题(JTZL2205);四川省无人系统智能感知控制技术工程实验室开放课题(WRXT2022-001);云基物联网高速公路建养设备智能化实验室开放课题(KF_2022_301002)。

摘　　要：语义分割任务是很多计算机视觉任务的前提与基础,在虚拟现实、无人驾驶等领域具有重要的应用价值。随着深度学习技术的快速发展,尤其是卷积神经网络(convolutional neural network,CNN)的出现,使得图像语义分割取得了长足的进步。首先,本文介绍了语义分割概念、相关背景和语义分割基本处理流程。然后,总结开源的2D、2.5D、3D数据集和其相适应的分割方法,详细描述了不同网络的分割特点、优缺点及分割精确度,得出监督学习是有效的训练方式。同时,介绍了权威的算法性能评价指标,根据不同方法的侧重点,对各个分割方法的相关实验进行了对比分析,指出了目前实验方面整体存在的问题,其中,DeepLab-V3+网络在分割精确度和速度方面都具有良好的性能,应用价值较高。在此基础上,本文针对国内外的研究现状,提出了当前面临的几点挑战和未来可能的研究方向。通过总结与分析,能够为相关研究人员进行图像语义分割相关研究提供参考。Introduced by Ohta in 1980,image semantic segmentation assigns each pixel in an image with a pre-defined label that represents its semantic category.Aiming to understand the different scenes of images,image semantic segmenta⁃tion has received much research attention in the field of computer vision.In recent years,many research laboratories around the world have carried out research work on image semantic segmentation based on deep learning.Academic confer⁃ences in the fields of automation,artificial intelligence,and pattern recognition also reported research results on semantic segmentation.At the same time,semantic segmentation serves as the premise and basis of many computer vision tasks and has important application value in virtual reality,such as automatic driving and human-computer interaction.With the rapid development of deep learning technology,especially the emergence of convolutional neural networks,image semantic segmentation technology has made great progress and has far outperformed traditional methods in terms of accuracy and efficiency.First,this paper introduces the concept of semantic segmentation along with its background and basic process.In general,image semantic segmentation based on deep learning goes through three processing modules,namely,the feature extraction,semantic segmentation,and refinement processing modules.Second,this paper summarizes the open source 2D,RGB-D,and 3D datasets that have been used in recent years and their corresponding segmentation methods.The semantic segmentation methods for 2D data are divided into method based on candidate region,method based on fully supervised learning,and method based on weakly supervised learning.As RGB-D and 3D date,only a few semantic seg⁃mentation methods need to be classified,thus no further classification is performed.This paper describes in detail the net⁃work structure of several classical algorithms,the segmentation characteristics,advantages,and disadvantages of different networks,and their segmentation accuracy.Through th

关 键 词：深度学习 图像语义分割(ISS) 卷积神经网络(CNN) 监督学习 DeepLab-V3+网络 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]