董彦锋,胡伍生,余龙飞,龙凤阳,张良

• 1:东南大学交通学院
• 2:同方威视技术股份有限公司

摘要(Abstract)：

针对城市行道树调查中,街景影像背景环境复杂多变、行道树个体差异大,依靠目视判读费时费力的问题,该文基于车载移动测量系统采集的全景影像数据,利用深度学习算法,在快速区域卷积神经网络的目标检测方法基础上,建立适用于街景行道树检测的深度神经网络模型。模型采用基于共有显著性区域及冗余策略的行道树多示例目标候选区域选择方法,使用车载图像的几何约束进一步筛选合适的候选区域,从而实现行道树目标候选区域的统一选择,提升行道树目标的检测效果。实验结果表明,该文提出的方法能够实现多种行道树的准确自动识别与提取,进而大大降低行道树绿化调查的成本。

关键词(KeyWords)： 卷积神经网络;候选区域选择;目标检测;行道树

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目（41574022）;; 江苏省研究生科研与实践创新计划项目（KYCX17_0150）

作者(Author): 董彦锋,胡伍生,余龙飞,龙凤阳,张良

DOI: 10.16251/j.cnki.1009-2307.2021.02.020

参考文献(References)：

• [1] 李玉兵,王鹏.车载激光点云数据中行道树三维信息自动提取研究[J].测绘与空间地理信息,2018,41(8):40-43.(LI Yubing,WANG Peng.Automatic extraction of street tree 3D information based on vehicle-borne LiDAR point cloud[J].测绘与空间地理信息,2018,41(8):40-43.)
• [2] 于圣坤,杨梅.胶州市城区行道树绿化调查与分析[J].黑龙江农业科学,2011(8):59-62.(YU Shengkun,YANG Mei.Investigation and analysis of street trees greening in Jiaozhou city[J].Heilongjiang Agricultural Sciences,2011(8):59-62.)
• [3] 吴宾,余柏蒗,岳文辉,等.一种基于车载激光扫描点云数据的单株行道树信息提取方法[J].华东师范大学学报(自然科学版),2013(2):38-49.(WU Bin,YU Bailang,YUE Wenhui,et al.Method for identifying individual street trees from the cloud data of the vehicle-borne laser scanning points[J].Journal of East China Normal University(Natural Science),2013(2):38-49.)
• [4] 李艳红.车载移动测量系统数据配准与分类识别关键技术研究[D].武汉:武汉大学,2014.(LI Yanhong.Research of data registration and classification for vehicle-borne mobile measurement system[D].Wuhan:Wuhan University,2014.)
• [5] 尹宝才,王文通,王立春.深度学习研究综述[J].北京工业大学学报,2015,41(1):48-59.(YIN Baocai,WANG Wentong,WANG Lichun.Review of deep learning[J].Journal of Beijing University of Technology,2015,41(1):48-59.)
• [6] SERMANET P,EIGEN D,ZHANG X,et al.OverFeat:integrated recognition,localization and detection using convolutional networks[J/OL].Computer Vision and Pattern Recognition,2013[2019-09-26].https://arxiv.org/pdf/1312.6229.pdf.
• [7] GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Washington,DC:IEEE Computer Society,2014.
• [8] HE K,ZHANG X,REN S,et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2014,37(9):1904-1916.
• [9] GIRSHICK R.Fast R-CNN[C]//IEEE International Conference on Computer Vision.[S.l.]:[s.n.],2015.
• [10] REN S,HE K,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
• [11] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Washington,DC:IEEE Computer Society,2016.
• [12] LIU W,ANGUELOV D,ERHAN D,et al.SSD:single shot multibox detector[C]//European Conference on Computer Vision.Cham:Springer,2016.
• [13] REDMON J,FARHADI A.YOLO9000:better,faster,stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).[S.l.]:[s.n.],2017:6517-6525.
• [14] 冯赟龙,刘勇,何王全.一种基于深度学习的性能分析框架设计与实现[J].计算机工程与科学,2018,40(6):984-991.(FENG Yunlong,LIU Yong,HE Wangquan.A deep learning based program performance analysis framework[J].Computer Engineering and Science,2018,40(6):984-991.)
• [15] 李想.深度卷积网络预训练策略的研究与应用[D].大连:大连理工大学,2017.(LI Xiang.Research and application of the pretraining strategies of deep convolutional neural netowrk[D].Dalian:Dalian University of Technology,2017.)
• [16] 吕燕彬.深度神经网络架构改进和训练性能提升的研究[D].太原:中北大学,2016.(LYU Yanbin.The research on improving of architecture and trainning performance of deep neural network[D].Taiyuan:North University of China,2016.)