|本期目录/Table of Contents|

[1]张瑞林,张俊为,桂江生,等.基于改进YOLOv2网络的遗留物检测算法[J].浙江理工大学学报,2018,39-40(自科3):325-332.
 ZHANG Ruilin,ZHANG Junwei,GUI Jiangsheng,et al.Abandoned object detection algorithm based on improved of YOLOv2 network[J].Journal of Zhejiang Sci-Tech University,2018,39-40(自科3):325-332.
点击复制

基于改进YOLOv2网络的遗留物检测算法()
分享到:

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
第39-40卷
期数:
2018年自科3期
页码:
325-332
栏目:
出版日期:
2018-05-19

文章信息/Info

Title:
Abandoned object detection algorithm based on improved of YOLOv2 network
文章编号:
1673-3851 (2018) 05-0325-08
作者:
张瑞林张俊为桂江生高春波包晓安
浙江理工大学信息学院,杭州 310018
Author(s):
ZHANG Ruilin ZHANG Junwei GUI Jiangsheng GAO Chunbo BAO Xiaoan
School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China
关键词:
YOLOv2网络遗留物检测残差网络
分类号:
TS195.644
文献标志码:
A
摘要:
为了提高在复杂环境下检测遗留物体的准确度和实时性,提出了一种基于改进YOLOv2网络的遗留物检测算法。该算法在YOLOv2网络结构基础上结合残差网络,将浅层和深层特征多次融合,在基本不增加原有模型计算量和时间的情况下,提高了监控画面中检测小体积遗留物体的性能;同时以YOLOv2目标检测为基础,排除驻留行人和动物等非物体目标的干扰,并对目标筛选得到的可疑目标跟踪计时,停留时间超过阈值的目标标记为遗留物。以PETS2006和iLIDS作为数据集进行实验,结果表明:该算法在提高遗留物检测准确度的同时缩短了处理时间,对人流密集的复杂环境抗干扰能力强。

参考文献/References:

[1] Muchtar K, Lin C Y, Kang L W, et al. Abandoned object detection in complicated environments[C]//Signal and Information Processing Association Summit and Conference. IEEE,2014:1-6.
[2] 张超,吴小培,周建英,等.基于改进混合高斯建模和短时稳定度的遗留物检测算法[J].信号处理,2012,28(8):1101-1111.
[3] Tripathi R K, Jalal A S, Bhatnagar C. A framework for abandoned object detection from video surveillance[C]//Computer Vision, Pattern Recognition, Image Processing and Graphics. IEEE,2014:1-4.
[4] 叶立仁,何盛鸿,赵连超.复杂环境下的遗留物检测算法[J].计算机工程与科学,2015,37(5):986-992.
[5] Lin K, Chen S C, Chen C S, et al. Abandoned object detection via temporal consistency modeling and backtracing verification for visual surveillance[J]. IEEE Transactions on Information Forensics & Security,2015,10(7):1359-1370.
[6] Lecun Y, Bengio Y, Hinton G. Deep learning[J]. Nature,2015,521(7553):436-444.
[7] Redmon J, Divvala S, Girshick R, et al. You only look once:Unified, realtime object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE,2016:779-788.
[8] 魏湧明,全吉成,侯宇青阳.基于YOLOv2的无人机航拍图像定位研究[J].激光与光电子学进展,2017,54(11):111002.
[9] Huang W, Jin L, Luo Y, et al. A novel algorithm for abandoned object detection[C]//IEEE International Conference on Information and Automation. IEEE,2017:1583-1587.
[10] Bhargava M, Chen C C, Ryoo M S, et al. Detection of abandoned objects in crowded environments[C]//Advanced Video and Signal Based Surveillance, 2007, IEEE Conference on. IEEE,2007:271-276.

备注/Memo

备注/Memo:
收稿日期: 2017-10-13
网络出版日期: 2017-12-11
基金项目: 国家自然科学基金项目(61502430,61379036,61562015);浙江理工大学521人才培养计划
作者简介: 张瑞林(1961-),男,浙江嵊州人,教授,博士,主要从事计算机应用技术等方面的研究
通信作者: 包晓安,E-mail:baoxiaoan@zstu.edu.cn
更新日期/Last Update: 2018-06-19