基于自适应多尺度特征融合网络的车辆检测方法

申铉京; 李涵宇; 黄永平; 王玉

doi:10.12263/DZXB.20220281

您当前的位置：

首页 >

文章列表页 >

基于自适应多尺度特征融合网络的车辆检测方法

更新时间：2023-03-23

- 基于自适应多尺度特征融合网络的车辆检测方法
- A Vehicle Detection Method Based on Adaptive Multi-Scale Feature Fusion Network
- 电子学报 2023年页码：1-9
- 作者机构：
  
  1.吉林大学计算机科学与技术学院，吉林长春 130012
  2.吉林大学软件学院，吉林长春 130012
- 作者简介：
  
  [ "申铉京男，1958年出生，吉林和龙人.吉林大学计算机科学与技术学院教授.主要研究方向为多媒体技术，计算机图像处理，智能测量系统，光电混合系统.E-mail: xjshen@jlu.edu.cn" ]
  [ "李涵宇男，1997年出生，江苏扬州人.吉林大学软件学院硕士研究生.主要研究方向为图像处理与模式识别.E-mail: hanyul19@mails.jlu.edu.cn" ]
  [ "黄永平男，1964年出生，吉林白城人.吉林大学计算机科学与技术学院副教授.主要研究方向为嵌入式软件与信息物理融合系统.E-mail: hyp@jlu.edu.cn" ]
  [ "王玉（通讯作者）男，1983年出生，黑龙江双鸭山人.吉林大学计算机科学与技术学院副教授.主要研究方向包括多媒体技术、机器学习与智能系统.E-mail: wangyu001@jlu.edu.cn" ]
- 基金信息：
  
  吉林省科技发展计划项目(20180201064SF)
- DOI：10.12263/DZXB.20220281
  中图分类号： TP391;
- 收稿：2022-03-18，
  
  修回：2022-09-08，
  
  网络出版：2023-03-23，
- 稿件说明：
移动端阅览
申铉京, 李涵宇, 黄永平, 等. 基于自适应多尺度特征融合网络的车辆检测方法[J/OL]. 电子学报, 2023,1-9.

SHEN Xuan-jing, LI Han-yu, HUANG Yong-ping, et al. A Vehicle Detection Method Based on Adaptive Multi-Scale Feature Fusion Network[J/OL]. ACTA ELECTRONICA SINICA, 2023, 1-9.
申铉京, 李涵宇, 黄永平, 等. 基于自适应多尺度特征融合网络的车辆检测方法[J/OL]. 电子学报, 2023,1-9. DOI： 10.12263/DZXB.20220281.

SHEN Xuan-jing, LI Han-yu, HUANG Yong-ping, et al. A Vehicle Detection Method Based on Adaptive Multi-Scale Feature Fusion Network[J/OL]. ACTA ELECTRONICA SINICA, 2023, 1-9. DOI： 10.12263/DZXB.20220281.

摘要

为了提高车辆检测精度，解决小目标车辆难以检测的问题，本文提出了自适应多尺度特征融合网络（Adaptive Multi-scale Feature Fusion Network，AMFFN），并基于该网络对YOLO v4进行了改进，取得了更好的检测效果.该网络通过使用多个空间金字塔池化，提高特征的表示能力.提出的AMFFN跨层融合了多尺度的特征，并为不同尺度的特征层分配可学习的权重.为了更好地获得特征的细节信息，本文选择了DY-ReLU作为激活函数，它可以随输入动态变化.AMFFN可以被视为一个可重用的模块，通过反复融合特性来获得更精细的特性.为了避免复杂的网络结构导致的巨大参数量，使用深度可分离卷积替换普通卷积，以降低参数量，提高网络检测速度.实验结果表明，本文提出的方法相比YOLO v4提高了1.90%的AP，检测速度提高了5 FPS.

Abstract

In order to improve the vehicle detection accuracy and solve the problem that small vehicles are difficult to detect

an adaptive multi-scale feature fusion network (AMFFN) is proposed and better performance is achieved after applying it to YOLO v4. To improve the representation capability of features

spatial pyramid pooling modules are employed on each feature map. The proposed AMFFN fuses features of multiple scales across layers and assigns learnable weights to layers of different scales. In order to achieve detailed information better

we select DY-ReLU as activation function

which can change dynamically with input. AMFFN can be treated as a reusable module to obtain more refined features by repeatedly fusing features. To avoid the huge amount of parameters caused by the complex network

depthwise separable convolution is used to replace the normal convolution in order to reduce the amount of parameters and increase the speed of detection. Experimental results show that the proposed method improves the AP by 1.90% and the detection speed by 5 FPS.

关键词

Keywords

references

PIAO J , MCDONALD M . Advanced driver assistance systems from autonomous to cooperative approach [J]. Transport Reviews , 2008 , 28 ( 5 ): 659 - 684 .

ANURADHA R . RETRACTED ARTICLE: Feature selection and classification methods for vehicle tracking and detection [J]. Journal of Ambient Intelligence and Humanized Computing , 2021 , 12 ( 3 ): 4269 - 4279 .

LI P L , CHEN X Z , SHEN S J . Stereo R-CNN based 3D object detection for autonomous driving [C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Long Beach : IEEE , 2020 : 7636 - 7644 .

WANG L , LU Y , WANG H , et al . Evolving boxes for fast vehicle detection [C]// 2017 IEEE International Conference on Multimedia and Expo (ICME) . Hong Kong : IEEE , 2017 : 1135 - 1140 .

YAN Z Y , YUAN Y C , ZUO W M , et al . Perspective-guided convolution networks for crowd counting [C]// 2019 IEEE/CVF International Conference on Computer Vision (ICCV) . Seoul : IEEE , 2020 : 952 - 961 .

SRIVASTAVA S , . Video-based real-time surveillance of vehicles [J]. Journal of Electronic Imaging , 2013 , 22 ( 4 ): 041103 .

DALAL N , TRIGGS B . Histograms of oriented gradients for human detection [C]// 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition . San Diego : IEEE , 2005 : 886 - 893 .

MITA T , KANEKO T , HORI O . Joint haar-like features for face detection [C]// Tenth IEEE International Conference on Computer Vision (ICCV'05) . Beijing : IEEE , 2005 : 1619 - 1626 .

SÁNCHEZ A V D . Advanced support vector machines and kernel methods [J]. Neurocomputing , 2003 , 55 ( 1/2 ): 5 - 20 .

BAEK Y M , KIM W Y . Forward vehicle detection using cluster-based AdaBoost [J]. Optical Engineering , 2014 , 53 ( 10 ): 102103 .

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 . Columbus : IEEE , 2014 : 580 - 587 .

GIRSHICK R . Fast R-CNN [C]// 2015 IEEE International Conference on Computer Vision (ICCV) . Santiago : IEEE , 2016 : 1440 - 1448 .

REN S Q , HE K M , 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 , 2017 , 39 ( 6 ): 1137 - 1149 .

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) . Las Vegas : IEEE , 2016 : 779 - 788 .

REDMON J , FARHADI A . YOLO9000: Better, faster, stronger [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Honolulu : IEEE , 2017 : 6517 - 6525 .

REDMON J , FARHADI A . YOLOv3: An incremental improvement [EB/OL]. ( 2018-04-08 )[ 2022-03 ]. https://arxiv.org/abs/1804.02767 https://arxiv.org/abs/1804.02767 .

BOCHKOVSKIY A , WANG C Y , LIAO H Y M . YOLOv4: Optimal speed and accuracy of object detection [EB/OL]. ( 2020-04-23 )[ 2023-03-13 ]. https://arxiv.org/abs/2004.10934 https://arxiv.org/abs/2004.10934 .

GLENN J . YOLOV 5 . ( 2021 )[ 2022-03 ]. https://github.com/ultralytics/yolov5 https://github.com/ultralytics/yolov5 .

LIU W , ANGUELOV D , ERHAN D , et al . SSD: Single shot multibox detector [C]// European Conference on Computer Vision - ECCV 2016 . Amsterdam : Springer , 2016 : 21 - 37 .

CHEN Y P , DAI X Y , LIU M C , et al . Dynamic ReLU [C]// European Conference on Computer Vision - ECCV 2020 . Glasgow : Springer , 2020 : 351 - 367 .

ZHOU Y , LIU L , SHAO L , et al . Fast automatic vehicle annotation for urban traffic surveillance [J]. IEEE Transactions on Intelligent Transportation Systems , 2018 , 19 ( 6 ): 1973 - 1984 .

YUAN X , SU S , CHEN H J . A graph-based vehicle proposal location and detection algorithm [J]. IEEE Transactions on Intelligent Transportation Systems , 2017 , 18 ( 12 ): 3282 - 3289 .

CHEN W P , QIAO Y T , LI Y J . Inception-SSD: An improved single shot detector for vehicle detection [J]. Journal of Ambient Intelligence and Humanized Computing , 2022 , 13 ( 11 ): 5047 - 5053 .

ZHANG S L , CHAI L , JIN L Z . Vehicle detection in UAV aerial images based on improved YOLOv3 [C]// 2020 IEEE International Conference on Networking, Sensing and Control (ICNSC) . Nanjing : IEEE , 2020 : 1 - 6 .

CHEN L , DING Q W , ZOU Q , et al . DenseLightNet: A light-weight vehicle detection network for autonomous driving [J]. IEEE Transactions on Industrial Electronics , 2020 , 67 ( 12 ): 10600 - 10609 .

JAMIYA S S , RANI E P . LittleYOLO-SPP: A delicate real-time vehicle detection algorithm [J]. Optik , 2021 , 225 : 165818 .

WANG C Y , MARK LIAO H Y , WU Y H , et al . CSPNet: A new backbone that can enhance learning capability of CNN [C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) . Seattle : IEEE , 2020 : 1571 - 1580 .

HOWARD A G , ZHU M L , CHEN B , et al . MobileNets: Efficient convolutional neural networks for mobile vision applications [EB/OL]. ( 2017-04-17 )[ 2022-03 ]. https://arxiv.org/abs/1704.04861 https://arxiv.org/abs/1704.04861 .

LONGYIN , WEN , . UA-DETRAC: A new benchmark and protocol for multi-object detection and tracking [J]. Computer Vision and Image Understanding , 2020 , 193 : 102907 .

TAN M X , PANG R M , LE Q V . EfficientDet: scalable and efficient object detection [C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Seattle : IEEE , 2020 : 10778 - 10787 .

LIN T Y , GOYAL P , GIRSHICK R , et al . Focal loss for dense object detection [C]// 2017 IEEE International Conference on Computer Vision (ICCV) . Venice : IEEE , 2017 : 2999 - 3007 .

ZHOU X Y , WANG D Q , KRÄHENBÜHL P . Objects as points [EB/OL]. ( 2019-04-16 )[ 2022-03 ]. https://arxiv.org/abs/1904.07850 https://arxiv.org/abs/1904.07850 .

王玉 , 李涵宇 , 申铉京 , 等 . 一种基于多层特征融合的车辆检测方法 : CN113420706B [P]. 2022-05-24 .

WANG Y , LI H Y , SHEN X J , et al . Vehicle detection method based on multilayer feature fusion : CN113420706B [P]. 2022-05-24 . (in Chinese) .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种基于改进CRNN的轻量化乐谱识别方法

基于级联式逆残差网络的遥感图像轻量目标检测算法

基于并行附加特征提取网络的SSD地面小目标检测模型