Lightweight pedestrian detection for multiple scenes

Yunzuo ZHANG; Wenbo LI; Wei GUO; Zhouchen SONG

doi:10.37188/OPE.20223014.1764

您当前的位置：

首页 >

文章列表页 >

Lightweight pedestrian detection for multiple scenes

Information Sciences | 更新时间：2022-08-08

- Lightweight pedestrian detection for multiple scenes
- Optics and Precision Engineering Vol. 30, Issue 14, Pages: 1764-1774(2022)
- 作者机构：
  
  石家庄铁道大学信息科学与技术学院，河北石家庄 050043
- 作者简介：
- 基金信息：
- DOI：10.37188/OPE.20223014.1764
  CLC： TP391.4
- Received：19 April 2022，
  
  Revised：27 April 2022，
  
  Published：25 July 2022
- 稿件说明：
移动端阅览
张云佐,李文博,郭威等.面向多元场景的轻量级行人检测[J].光学精密工程,2022,30(14):1764-1774.

ZHANG Yunzuo,LI Wenbo,GUO Wei,et al.Lightweight pedestrian detection for multiple scenes[J].Optics and Precision Engineering,2022,30(14):1764-1774.
张云佐,李文博,郭威等.面向多元场景的轻量级行人检测[J].光学精密工程,2022,30(14):1764-1774. DOI： 10.37188/OPE.20223014.1764.

ZHANG Yunzuo,LI Wenbo,GUO Wei,et al.Lightweight pedestrian detection for multiple scenes[J].Optics and Precision Engineering,2022,30(14):1764-1774. DOI： 10.37188/OPE.20223014.1764.

摘要

多元场景中行人检测是当前计算机视觉领域的研究热点，尽管备受关注的深度学习能够提供很高的检测精度，但随之而来的高复杂度运算严重限制了其在可移动平台上的部署。为此，本文提出了一种面向多元场景的轻量级行人检测算法。该算法首先构建深、浅层特征融合网络以学习多尺度行人的纹理特性；然后设计了跨维特征引导注意力模块，用于保留特征提取过程中通道间、空间内的交互信息。最后基于剪枝策略去除模型中的冗余通道，以降低算法复杂度。此外，本文还设计了自适应Gamma矫正算法，以消减多元场景下光照、阴影等外界干扰对检测结果的影响。实验结果表明，本文所提方法在检测精度相当的条件下，能将模型大小压缩至10 MB，处理速度可达93 Frame/s，明显优于当前主流方法。

Abstract

Currently， pedestrian detection in multiple scenes is a research hotspot in the field of computer vision. Deep learning has attracted considerable attention and can provide high detection accuracy； however， the subsequent high-complexity operations seriously limit its application on mobile platforms. To address this problem， this paper proposes a lightweight pedestrian detection algorithm for multiple scenes. Firstly， a deep and shallow feature fusion network is constructed to learn the texture features of multi-scale pedestrians. Secondly， a cross-dimensional feature-guided attention module is designed to retain the interactive information between channels and spaces in the process of feature extraction. Finally， the redundant channels in the model are trimmed using the pruning strategy， to reduce the algorithm complexity. In addition， an adaptive Gamma correction algorithm is designed to reduce the influence of external disturbances， such as illumination and shadows， on the detection results of multiple scenes. The experimental results show that the proposed method can compress the model volume to 10 MB， and the processing speed can reach 93 Frame/s while achieving similar detection accuracy， which outperforms the current mainstream methods.

关键词

Keywords

references

YADAV R P ， SENTHAMILARASU V ， KUTTY K ， et al . Implementation of robust HOG-SVM based pedestrian classification ［J］. International Journal of Computer Applications ， 2015 ， 114 （ 19 ）： 10 - 16 . doi: 10.5120/20084-2026 http://dx.doi.org/10.5120/20084-2026

SABZMEYDANI P ， MORI G . Detecting pedestrians by learning shapelet features ［C］. 2007 IEEE Conference on Computer Vision and Pattern Recognition . 1722，2007 ， Minneapolis ， MN ， USA . IEEE ， 2007 ： 1 - 8 . doi: 10.1109/cvpr.2007.383134 http://dx.doi.org/10.1109/cvpr.2007.383134

CAO J ， PANG Y ， XIE J ， et al . From handcrafted to deep features for pedestrian detection： a survey ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence ， 2021 . doi: 10.1109/tpami.2021.3076733 http://dx.doi.org/10.1109/tpami.2021.3076733

CAI Z ， FAN Q ， FERIS R S ， et al . A unified multi-scale deep convolutional neural network for fast object detection ［C］. European conference on computer Vision. Springer ， Cham ， 2016 ： 354 - 370 . doi: 10.1007/978-3-319-46493-0_22 http://dx.doi.org/10.1007/978-3-319-46493-0_22

YANG P Y ， ZHANG G F ， WANG L ， et al . A part-aware multi-scale fully convolutional network for pedestrian detection ［J］. IEEE Transactions on Intelligent Transportation Systems ， 2021 ， 22 （ 2 ）： 1125 - 1137 . doi: 10.1109/tits.2019.2963700 http://dx.doi.org/10.1109/tits.2019.2963700

邹梓吟，盖绍彦，达飞鹏，等 . 基于注意力机制的遮挡行人检测算法［J］. 光学学报， 2021 ， 41 （ 15 ）： 1515001 . doi: 10.3788/aos202141.1515001 http://dx.doi.org/10.3788/aos202141.1515001

ZOU Z Y ， GAI S Y ， DA F P ， et al . Occluded pedestrian detection algorithm based on attention mechanism ［J］. Acta Optica Sinica ， 2021 ， 41 （ 15 ）： 1515001 . （in Chinese） . doi: 10.3788/aos202141.1515001 http://dx.doi.org/10.3788/aos202141.1515001

WU J L ， ZHOU C L ， YANG M ， et al . Temporal-context enhanced detection of heavily occluded pedestrians ［C］. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. 1319，2020 ， Seattle， WA， USA. IEEE ， 2020 ： 13427 - 13436 . doi: 10.1109/cvpr42600.2020.01344 http://dx.doi.org/10.1109/cvpr42600.2020.01344

时小虎，吴佳琦，吴春国，等 . 基于残差网络的弯道增强车道线检测方法［J/OL］. 吉林大学学报（工学版）： 1 - 9 . ［ 2022-04-02 ］. DOI： 10.13229/j.cnki.jdxbgxb20210618 http://dx.doi.org/10.13229/j.cnki.jdxbgxb20210618 .

SHI X H ， WU J Q ， WU C G ， et al . Residual network based curve enhanced lane detection method ［J/OL］. Journal of Jilin University（Engineering and Technology Edition）： 1 - 9 . ［ 2022-04-02 ］. DOI： 10.13229/j.cnki.jdxbgxb20210618. http://dx.doi.org/10.13229/j.cnki.jdxbgxb20210618. （in Chinese）

GE Z ， LIU S ， WANG F ， et al . Yolox： Exceeding yolo series in 2021 ［J］. arXiv preprint arXiv ： 2017 ，08430， 2021 .

HOU Q B ， ZHOU D Q ， FENG J S . Coordinate attention for efficient mobile network design ［C］. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. 2025，2021 ， Nashville， TN， USA. IEEE ， 2021 ： 13708 - 13717 . doi: 10.1109/cvpr46437.2021.01350 http://dx.doi.org/10.1109/cvpr46437.2021.01350

LIU Z ， LI J G ， SHEN Z Q ， et al . Learning efficient convolutional networks through network slimming ［C］. 2017 IEEE International Conference on Computer Vision . 2229，2017 ， Venice， Italy . IEEE ， 2017 ： 2755 - 2763 . doi: 10.1109/iccv.2017.298 http://dx.doi.org/10.1109/iccv.2017.298

任彬，王宇庆，丛振，等 . 基于MPSOC的航空图像目标检测系统设计［J］. 液晶与显示， 2021 ， 36 （ 7 ）： 1006 - 1017 . doi: 10.37188/CJLCD.2020-0310 http://dx.doi.org/10.37188/CJLCD.2020-0310

REN B ， WANG Y Q ， CONG Z ， et al . Design of aerial image target detection system based on MPSOC ［J］. Chinese Journal of Liquid Crystals and Displays ， 2021 ， 36 （ 7 ）： 1006 - 1017 . （in Chinese） . doi: 10.37188/CJLCD.2020-0310 http://dx.doi.org/10.37188/CJLCD.2020-0310

周经美，王钰，宁航，等 . 面向多元场景结合GLNet的车道线检测算法［J］. 中国公路学报， 2021 ， 34 （ 7 ）： 118 - 127 . doi: 10.3969/j.issn.1001-7372.2021.07.010 http://dx.doi.org/10.3969/j.issn.1001-7372.2021.07.010

ZHOU J M ， WANG Y ， NING H ， et al . Lane detection algorithm based on GLNet for multiple scenes ［J］. China Journal of Highway and Transport ， 2021 ， 34 （ 7 ）： 118 - 127 . （in Chinese） . doi: 10.3969/j.issn.1001-7372.2021.07.010 http://dx.doi.org/10.3969/j.issn.1001-7372.2021.07.010

BABAKHANI P ， ZAREI P . Automatic gamma correction based on average of brightness ［J］. Advances in Computer Science： an International Journal ， 2015 ， 4 （ 6 ）： 156 - 159 .

WANG S G ， CHENG J ， LIU H J ， et al . PCN： part and context information for pedestrian detection with CNNs ［J］. arXiv preprint arXiv ： 1804.04483 ， 2018 . doi: 10.5244/c.31.34 http://dx.doi.org/10.5244/c.31.34

Views

618

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Dense pedestrian detection algorithm in multi-branch non-anchor frame network

Polarization computational imaging super-resolution reconstruction with lightweight attention cascading network

Multispectral pedestrian detection network under modal adaptive weight learning mechanism

Pedestrian intruding railway clearance classification algorithm based on improved deep convolutional network

Pedestrian detection based on tree-structured graphical model of the human body and hybrid particle swarm clustering

Related Author

LÜ Zhixuan

WEI Xia

HUANG Deqi

WANG Jie

XU Guoming

MA Jian

WANG Yong

LI Yi

Related Institution

School of Electrical Engineering， XinJiang University

School of Internet， Anhui University

National Engineering Research Center for Agro-Ecological Big Data Analysis & Application， Anhui University

Anhui Province Key Laboratory of Polarized Imaging Detecting Technology， Army Artillery and Air Defense Forces Academy of PLA

Institute of Intelligent Technology， Anhui Wenda University of Information Engineering

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰