计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230400101-7.doi: 10.11896/jsjkx.230400101

• 图像处理&多媒体技术 • 上一篇    下一篇

方向感知金字塔聚合网络的道路中心线提取

张小青1, 王青旺1, 瞿信1, 沈世全2, 吴长义3, 刘菊4   

  1. 1 昆明理工大学信息工程与自动化学院 昆明 650504
    2 昆明理工大学交通工程学院 昆明 650504
    3 昭通市建筑工程质量监督站 云南 昭通 657099
    4 用友网络科技股份有限公司流程制造开发四部 成都 610095
  • 发布日期:2024-06-06
  • 通讯作者: 王青旺(wangqingwang@kust.edu.cn)
  • 作者简介:(zhangxiaoqing@stu.kust.edu.cn)
  • 基金资助:
    国家自然科学基金青年科学基金(62201237);云南省基础研究计划项目(202101BE070001-008)

Direction-aware Pyramidal Aggregation Network for Road Centerline Extraction

ZHANG Xiaoqing1, WANG Qingwang1, QU Xin1, SHEN Shiquan2, WU Changyi3, LIU Ju4   

  1. 1 School of Information Engineering and Automation,Kunming University of Technology,Kunming 650504,China
    2 School of Traffic Engineering,Kunming University of Technology,Kunming 650504,China
    3 Zhaotong City Construction Quality Supervision Station,Zhaotong,Yunnan 657099,China
    4 Process Manufacturing Development IV,Yonyou Network Technology Co,Chengdu 610095,China
  • Published:2024-06-06
  • About author:ZHANG Xiaoqing,born in 1997,postgraduate.Her main research interests include computer vision and deep lear-ning.
    WANG Qingwang,born in 1990,Ph.D,professor.His main research interests include the collaborative interpretation of multi-source remote sensing data and deep learning.
  • Supported by:
    Young Scientists Fund of the National Natural Science Foundation of China(62201237) and Yunnan Fundamental Research Projects(202101BE070001-008).

PDF (PC)

257

摘要： 道路中心线作为抽象类,无明确显性特征,进而造成模型无法准确提取道路中心线。针对该问题,文中将道路中心线提取建模为语义分割任务,并根据道路中心线的线性结构特点提出了一种方向感知金字塔聚合网络(Direction-aware Pyramidal Aggregation Network,DAPANet)。首先,针对道路中心线的空间分布特性及结构特点,设计了方向感知模块(Direction-aware Module,DAM),在主干网络(ResNet18)最后输出的4个层级上分别使用4个方向感知层提取道路中心线的方向特征。然后,进一步设计融合多向性特征的金字塔聚合模块(Pyramid Aggregation Module,PAM),融合4个层级提取到的结构特征,得到更具有鲁棒性的道路中心线特征。最后,在无人机平台下采集的真实数据上进行了实验,实验结果显示所提出的DAPANet取得了84.7%的mIoU和98.6%的Precision,道路中心线的IoU达到77.28%,性能优于其他先进的对比方法,证明了所提方法对提取道路中心线的有效性。

关键词: 方向感知, 金字塔聚合, 语义分割, 道路中心线提取

Abstract: As an abstract class,road centerlines have no explicit features,which in turn causes the model fail to extract road centerlines accurately.To address this problem,this paper models road centerline extraction as a semantic segmentation task,and proposes a direction-aware pyramidal aggregation network(DAPANet) based on the spatial linear structure of road centerlines.Firstly,for the spatial distribution characteristics and structural features of road centerlines,this paper designs the direction-aware module(DAM) to extract the features of road centerlines using four direction-aware layers on each of the four layers of the final output of the backbone network(ResNet18).Then,it further designs the pyramid aggregation module(PAM) to fuse the structural features extracted from the four layers to obtain a more robust road centerline feature.Experiments are conducted on real data collected under the UAV platform,and the experimental results show that the proposed DAPANet achieves 84.7% of mIoU and 98.6% of Precision,in which the IoU of road centerline reaches 77.28%,outperforming other advanced comparative methods and proving the effectiveness of the proposed method.

Key words: Direction-awareness, Pyramid aggregation, Semantic segmentation, Road centerline extraction

中图分类号: 

  • TP389
[1]WU H B,QIU W M,WANG J,et al.Intelligent multi-track inspection robot inspection system for urban rail vehicles[J].Locomotive Electric Drive,2022(1):45-52.
[2]ZHU J Y,LI P,LI X,et al.Intelligent inspection robot positioning system for urban rail vehicles[J].Railway Transport and Economy,2023,45(1):130-136.
[3]ZHAO X T,HU K,LU X P,et al.A method for fine detection of geological hazards in mines by low altitude aerial photography with unmanned aerial vehicles[J].Survey and Mapping Science,2014,39(6):49-52.
[4]PAN L,GUO T T.Intelligent inspection application of naturalgas long-distance pipeline by drones[J].Chemical Management,2023,653(2):124-126.
[5]ZHU C,ZHU J,BU T,et al.Monitoring and Identification of Road Construction Safety Factors via UAV[J].Sensors,2022,22(22):8797.
[6]SHERSTJUK V,ZHARIKOVA M,SOKOL I.Forest fire monitoring system based on UAV team,remote sensing,and image processing[C]//Procedings of IEEE Conference on Data Stream Mining and Processing(DSMP).2018:590-594.
[7]MIAO Z,WANG B,SHI W,et al.A semi-automatic method for road centerline extraction from VHR images[J].IEEE Geoscience and Remote Sensing Letters,2014,11(11):1856-1860.
[8]CHENG G,WANG Y,XU S,et al.Automatic road detectionand centerline extraction via cascaded end-to-end convolutional neural network[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(6):3322-3337.
[9]SHAO Z,ZHOU Z,HUANG X,et al.MRENet:Simultaneousextraction of road surface and road centerline in complex urban scenes from very high-resolution images[J].Remote Sensing,2021,13(2):239.
[10]YU C,WANG J,PENG C,et al.Bisenet:Bilateral segmentation network for real-time semantic segmentation[C]//European Conference on Computer Vision(ECCV).2018:325-341.
[11]ZHAO H,QI X,SHEN X,et al.Icnet for real-time semantic segmentation on high-resolution images[C]//European Conference on Computer Vision(ECCV).2018:405-420.
[12]WANG Y,ZHOU Q,LIU J,et al.Lednet:A lightweight encoder-decoder network for real-time semantic segmentation[C]//International Conference on Image Processing(ICIP).IEEE,2019:1860-1864.
[13]SHI W,MIAO Z,DEBAYLE J.An integrated method for urban main-road centerline extraction from optical remotely sensed imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2013,52(6):3359-3372.
[14]DAI J,MA R,GONG L,et al.A model-driven-to-sample-driven method for rural road extraction[J].Remote Sensing,2021,13(8):1417.
[15]LIU R,MIAO Q,SONG J,et al.Multiscale road centerlines extraction from high-resolution aerial imagery[J].Neurocomputing,2019,329:384-396.
[16]LIAN R,HUANG L.DeepWindow:Sliding window based ondeep learning for road extraction from remote sensing images[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2020,13:1905-1916.
[17]DONG G,YAN Y,SHEN C,et al.Real-time high-performance semantic image segmentation of urban street scenes[J].IEEE Transactions on Intelligent Transportation Systems,2020,22(6):3258-3274.
[18]LI S,YAN Q,SHI W,et al.HoloParser:Holistic Visual Parsing for Real-time Semantic Segmentation in Autonomous Driving[J].IEEE Transactions on Instrumentation and Measurement,2023,72:1-15.
[19]ZOU W,HUA G,ZHUANG Y,et al.Real-time Passable Area Segmentation with Consumer RGB-D Cameras for the Visually Impaired[J].IEEE Transactions on Instrumentation and Measurement,2023,72:1-11.
[20]YU C,GAO C,WANG J,et al.Bisenet v2:Bilateral networkwith guided aggregation for real-time semantic segmentation[J].International Journal of Computer Vision,2021,129:3051-3068.
[21]FAB M,LAI S,HUANG J,et al.Rethinkingbisenet for real-time semantic segmentation[C]//Conference on Computer Vision and Pattern Recognition(CVPR).2021:9716-9725.
[22]PAN H,HONG Y,SUN W,et al.Deep dual-resolution net-works for real-time and accurate semantic segmentation of traffic scenes[J].IEEE Transactions on Intelligent Transportation Systems,2023,24(3):3348-3460.
[23]XU J,XIONG Z,BHATTACHARYYA S P.Pidnet:A real-time semantic segmentation network inspired from pid controller[J].arXiv:2206.02066,2022.
[24]ELHASSAN M A M,YANG C,HUANG C,et al.SPFNet:Subspace Pyramid Fusion Network for Semantic Segmentation[J].arXiv:2204.01278,2022.
[25]SHI M,SHEN J,YI Q,et al.LMFFNet:A well-balanced lightweight network for fast and accurate semantic segmentation[J].IEEE Transactions on Neural Networks and Learning Systems,2023,34(6):3205-3219.
[26]ELHASSAN M A M,YANG C,HUANG C,et al.SS-FPN:Scale-ware Strip Attention Guided Feature Pyramid Network for Real-time Semantic Segmentation[J].arXiv:2206.07298,2022.
[27]SUN T,DI Z,CHE P,et al.Leveraging crowdsourcedgps datafor road extraction from aerial imagery[C]//Conference on Computer Vision and Pattern Recognition(CVPR).2019:7509-7518.
[28]MEI J,LI R J,GAO W,et al.CoANet:Connectivity attentionnetwork for road extraction from satellite imagery[J].IEEE Transactions on Image Processing,2021,30:8540-8552.
[29]WANG J,CHEN K,XU R,et al.Carafe:Content-aware reassembly of features[C]//International Conference on Computer Vision(ICCV).2019:3007-3016.
[30]LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//International Conference on Computer Vision(ICCV).2017:2980-2988.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发