融合可变形卷积网络的鱼眼图像目标检测

doi:10.19678/j.issn.1000-3428.0057485

计算机工程 ›› 2021, Vol. 47 ›› Issue (4): 248-255. doi: 10.19678/j.issn.1000-3428.0057485

融合可变形卷积网络的鱼眼图像目标检测

包俊, 刘宏哲

北京联合大学北京市信息服务工程重点实验室, 北京 100101

收稿日期:2020-02-24 修回日期:2020-03-31 发布日期:2020-04-07
作者简介:包俊(1995-),男,硕士研究生,主研方向为计算机视觉、深度学习、目标检测;刘宏哲(通信作者),教授、博士。
基金资助:
国家自然科学基金（61871039）；北京市自然科学基金（4184088）；北京市属高校高水平教师队伍建设支持计划项目（IDHT20170511）；北京联合大学研究生科研创新项目（YZ2020K001）。

Object Detection in Fisheye Images Combining Deformable Convolutional Networks

BAO Jun, LIU Hongzhe

Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing 100101, China

Received:2020-02-24 Revised:2020-03-31 Published:2020-04-07

摘要/Abstract

摘要： 环视鱼眼图像具有目标形变大和图像失真的缺点，导致传统网络结构在对鱼眼图像进行目标检测时效果不佳。为解决环视鱼眼图像中由于目标几何畸变而导致的目标检测难度大的问题，提出一种基于可变形卷积网络的鱼眼图像目标检测方法。将Cascade_RCNN中固定的卷积层和池化层分别替换为可变形卷积层和可变形池化层，使用Resnet50网络提取候选区域以获得检测框，级联具有不同IoU阈值的检测网络进行检测框抑制。在公开鱼眼图像数据集SFU_VOC_360和本文所采集的真实道路场景鱼眼图像数据集上进行实验，结果表明，该方法在鱼眼图像目标检测中具有有效性，目标检测准确率高于Cascade_RCNN网络。

关键词: 鱼眼图像, 可变形卷积, 可变形池化, 目标检测, 环视系统

Abstract: Due to the significant target shape changes and distortions in bird's eye view fisheye images,the conventional network structures do not perform well in target detection for fisheye images.To address the geometric distortions of the target,which increases the difficulty of target detection in bird's eye view fisheye images,this paper proposes an object detection method in fisheye images based on deformable convolutional network.The fixed convolution layer and pooling layer in Cascade_RCNN is replaced by the deformable convolution layer and deformable pooling layer.Then the candidate region is extracted by using Resnet50 to obtain the detection box,which is suppressed by cascading the detection networks with different Intersection-over-Union(IoU) thresholds.Experiments are carried out on the open fisheye image dataset SFU _VOC _360 and the manually collected fisheye image dataset of real on-road driving scenes.The experimental results demonstrate the effectiveness of the proposed method for object detection in fisheye images.Its detection accuracy is higher than that of Cascade_RCNN.

Key words: fisheye image, deformable convolution, deformable pooling, object detection, bird's eye view system

中图分类号:

TP391

包俊, 刘宏哲. 融合可变形卷积网络的鱼眼图像目标检测[J]. 计算机工程, 2021, 47(4): 248-255.

BAO Jun, LIU Hongzhe. Object Detection in Fisheye Images Combining Deformable Convolutional Networks[J]. Computer Engineering, 2021, 47(4): 248-255.

http://www.ecice06.com/CN/Y2021/V47/I4/248

图/表 14

20210425171452

20210425171455

20210425171458

20210425171500

20210425171504

20210425171506

20210425171509

20210425171512

20210425171515

20210425171517

20210425171520

20210425171523

20210425171526

20210425171529

参考文献

[1] DENG Liuyuan,YANG Ming,QIAN Yeqiang,et al.CNN based semantic segmentation for urban traffic scenes using fisheye camera[C]//Proceedings of 2017 IEEE Intelligent Vehicles Symposium.Washington D.C.,USA:IEEE Press,2017:231-236.
[2] 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 and Machine Intelligence,2015,39(6):91-99.
[3] MINAEE S, BOYKOV Y Y, PORIKLI F, et al. Image segmentation using deep learning:a survey[EB/OL].[2020-01-05]. https://arxiv.org/pdf/2001.05566.pdf.
[4] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1916.
[5] CHEN X L,FANG H,LIN T Y.Microsoft COCO captions[EB/OL].[2020-01-05].https://arxiv.org/pdf/1504.00325.pdf.
[6] MARK E,LUC V G,CHRISTOPHER K W,et al.The PASCAL visual object classes challenge[EB/OL].[2020-01-05].http://lear.inrialpes.fr/SicilyWorkshop/sat%20am/Data%20Sets%20Talks/voc2006.pdf.
[7] MARIUS C,MOHAMED O,SEBASTIAN R,et al.The cityscapes dataset for semantic urban scene under-standing[C]//Proceedings of IEEE Conference on Com-puter Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:3213-3223.
[8] FU J,BAJI I V,VAUGHAN R G.Datasets for face and object detection in fisheye images[J].Data in Brief,2019(27):132-139.
[9] YOGAMANI S,HUGHES C,HORGAN J,et al.Woodscape:a multi-task,multi-camera fisheye dataset for autonomous driving[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2019:9308-9318.
[10] KHASANOVA R,FROSSARD P.Graph-based classification of omnidirectional images[C]//Proceedings of IEEE International Conference on Computer Vision Workshop.Washington D.C.,USA:IEEE Press,2017:869-878.
[11] JEON Y,KIM J.Active convolution:learning the shape of convolution for image classification[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recogni-tion.Washington D.C.,USA:IEEE Press,2017:4201-4209.
[12] SU Y C,GRAUMAN K.Learning spherical convolution for fast features from 360° imagery[EB/OL].[2020-01-05].https://papers.nips.cc/paper/2017/file/0c74b7f78409a4022a2c4c5a5ca3ee19-Paper.pdf.
[13] BAEK J Y,CHELU I V,IORDACHE L,et al.Scene understanding networks for autonomous driving based on around view monitoring system[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:961-968.
[14] TATENO K,NAVAB N,TOMBARI F.Distortion-aware convolutional filters for dense prediction in panoramic images[C]//Proceedings of European Conference on Com-puter Vision.Berlin,Germany:Springer,2018:707-722.
[15] GAO Qun,ZHU Jun,WANG Qianqian,et al.Research on the object detection algorithm based on fisheye image[J].Control and Information Technology,2019(3):43-47.(in Chinses)高群,朱均,王芊芊,等.基于鱼眼图像的目标检测算法研究[J].控制与信息技术,2019(3):43-47.
[16] YAHIAOUI M,RASHED H,MARIOTTI L,et al.FisheyeMODNet:moving object detection on surround-view cameras for autonomous driving[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2019:123-156.
[17] DAI Jifeng,QI Haozhi,XIONG Yuwen,et al.Deformable convolutional networks[C]//Proceedings of IEEE Inter-national Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2017:764-773.
[18] ZHU X,HU H,LIN S,et al.Deformable ConvNets V2:more deformable,better results[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2019:9308-9316.
[19] DENG Liuyuan,YANG Ming,LI Hao,et al.Restricted deformable convolution-based road scene semantic segmentation using surround view cameras[J].IEEE Transactions on Intelligent Transportation Systems,2019,21(10):4350-4362.
[20] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:770-778.
[21] CAI Z W,VASCONCELOS N.Cascade R-CNN:delving into high quality object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:6154-6162.
[22] REVAUD J, ALMAZÁN J, REZENDE R S, et al. Learning with average precision:training image retrieval with a listwise loss[C]//Proceedings of IEEE/CVF Interna-tional Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2019:5107-5116.

选择文件类型/文献管理软件名称

选择包含的内容

融合可变形卷积网络的鱼眼图像目标检测

Object Detection in Fisheye Images Combining Deformable Convolutional Networks

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	徐春波, 闫娟, 杨慧斌, 王博, 吴晗. 基于目标检测和语义分割的视觉SLAM算法[J]. 计算机工程, 2023, 49(8): 199-206, 214.
[2]	宋志娜, 李莎, 杨建明, 徐川. 基于特征与区域定位增强的遥感舰船目标检测[J]. 计算机工程, 2023, 49(8): 257-264.
[3]	刘俊豪, 王美林, 谢兴, 宋烨兴, 许莉花. 基于改进YOLOv5的皮革瑕疵检测算法[J]. 计算机工程, 2023, 49(8): 240-249.
[4]	李强龙, 周新文, 位梦恩, 甘阳洲. 基于条形池化和注意力机制的街道场景红外目标检测算法[J]. 计算机工程, 2023, 49(8): 310-320.
[5]	闫兴亚, 匡娅茜, 白光睿, 李月. 基于深度学习的学生课堂行为识别方法[J]. 计算机工程, 2023, 49(7): 251-258.
[6]	聂志勇, 阴宇薇, 汤佳欣, 涂志刚. 一种基于边界框关键点距离的框回归算法[J]. 计算机工程, 2023, 49(7): 65-75.
[7]	李军侠, 王星驰, 殷梓, 石德硕. 边缘深度挖掘的弱监督显著性目标检测[J]. 计算机工程, 2023, 49(7): 169-178.
[8]	吴珊, 周凤. 基于改进SSD算法的小目标检测[J]. 计算机工程, 2023, 49(7): 179-188.
[9]	齐咏生, 杜晓旭, 朱俊峰, 高胜利, 刘利强. 基于增强型轻量深度网络的牧区牲畜高效检测[J]. 计算机工程, 2023, 49(7): 278-287.
[10]	谌雨章, 黄逸姿, 张钧涵. 基于多速率空洞卷积的多尺度水下小目标检测[J]. 计算机工程, 2023, 49(6): 257-264.
[11]	罗华峰, 沈奕菲, 阮黎翔, 杜奇伟, 郑翔, 陈智麒, 张胜. 边缘环境下面向实时目标检测的帧卸载调度算法[J]. 计算机工程, 2023, 49(5): 295-301,309.
[12]	王璐璐, 陈东方, 王晓峰. 一种基于锚框质量分布的动态标签分配策略[J]. 计算机工程, 2023, 49(4): 85-91,100.
[13]	宋鹏鹏, 龚声蓉, 钟珊, 周立凡, 凤黄浩. 基于双注意力擦除和注意力信息聚合的弱监督目标检测[J]. 计算机工程, 2023, 49(3): 113-120,127.
[14]	唐榕, 李骞, 唐绍恩. 基于多目标的能见度检测方法[J]. 计算机工程, 2023, 49(2): 314-320.
[15]	王朕, 李豪, 严冬梅, 竺永荣. 基于改进YOLOv5的路面病害检测模型[J]. 计算机工程, 2023, 49(2): 15-23.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

融合可变形卷积网络的鱼眼图像目标检测

Object Detection in Fisheye Images Combining Deformable Convolutional Networks

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献

相关文章 15

编辑推荐

Metrics

本文评价