基于深度卷积神经网络的运动目标检测方法

doi:10.16182/j.issn1004731x.joss.19-FZ0368

系统仿真学报 ›› 2019, Vol. 31 ›› Issue (11): 2275-2280.doi: 10.16182/j.issn1004731x.joss.19-FZ0368

基于深度卷积神经网络的运动目标检测方法

卢裕秋, 孙金玉, 马世伟^*

上海大学机电工程与自动化学院,上海 200444

收稿日期:2019-06-20 修回日期:2019-07-23 出版日期:2019-11-10 发布日期:2019-12-13
作者简介:卢裕秋(1994-),男,江苏宿迁,硕士生,研究方向为模式识别;孙金玉(1991-),女,江苏连云港,博士生,研究方向为图像处理;马世伟(通讯作者1965-),男,甘肃嘉峪关,博士,教授,研究方向为信号处理、模式识别。
基金资助:
新疆兵团重大项目子项目(2018AA008-04)

Moving Object Detection Based on Deep Convolutional Neural Network

Lu Yuqiu, Sun Jinyu, Ma Shiwei^*

School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China

Received:2019-06-20 Revised:2019-07-23 Online:2019-11-10 Published:2019-12-13

摘要/Abstract

摘要： 针对运动目标检测中的间歇性运动问题,设计了一个深度卷积神经网络MONet。在缺乏训练数据集的情况下,利用仿射变换生成一个合成数据集GoChairs,并在此基础上进行网络的训练和测试。结果表明,训练后的MONet能够有效地根据像素点之间的对应关系检测出运动的目标。传统的运动目标检测数据集CDnet和I2R被用于测试以验证该网络的泛化性能。针对目标的间歇性运动问题,MONet与经典方法进行了定性和定量的比较。实验结果证明了该网络在检测间歇性运动的目标时的优越性。

关键词: 运动目标检测, 间歇性运动, 深度卷积神经网络, 合成数据集

Abstract: A deep convolution neural network MONet is designed for the intermittent motion problem in moving objects detection. In the absence of training data sets, a synthetic dataset GoChairs is generated by affine transformation, and on this basis the network training and testing are performed. The results show that the trained MONet can effectively detect the moving objects based on the correspondence between the pixels. The traditional datasets CDnet and I2R are also tested to verify the generalization performance of the network. In addition, MONet is compared qualitatively and quantitatively with classical methods for the intermittent motion problem of the objects. The experimental results demonstrate the superiority of the network in detecting the objects with intermittent motion.

Key words: moving object detection, intermittent motion, deep convolutional neural network, synthetic dataset

中图分类号:

TP391.4

卢裕秋, 孙金玉, 马世伟. 基于深度卷积神经网络的运动目标检测方法[J]. 系统仿真学报, 2019, 31(11): 2275-2280.

Lu Yuqiu, Sun Jinyu, Ma Shiwei. Moving Object Detection Based on Deep Convolutional Neural Network[J]. Journal of System Simulation, 2019, 31(11): 2275-2280.

参考文献

[1] Pankaj D.Real-Time Surveillance for Critical Activity Detection in ICUs[C]. Proceedings of the Second International Conference on Computer and Communication Technologies. New Delhi: Springer, 2016.
[2] Tian B, Morris B T, Tang M, et al.Hierarchical and Networked Vehicle Surveillance in ITS: A Survey[J]. IEEE Transactions on Intelligent Transportation Systems (S1558-0016), 2017, 18(1): 25-48.
[3] Robertson N, Reid I.A general method for human activity recognition in video[J]. Computer Vision and Image Understanding (S1077-3142), 2006, 104(2/3): 232-248.
[4] Nait-Charif H, Mc Kenna S J. Head Tracking and Action Recognition in a Smart Meeting Room[J]. IEEE International Workshop on Performance Evaluation of Tracking and Surveillance (S2157-491X), 2003.
[5] C H Anderson, P J Burt, G S van der Wal. Change detection and tracking using pyramid transform techniques[C]. Cambridge: Proc. SPIE, 1985.
[6] Wren C, Azarbayejani A, Darrell T, et al.Pfinder: real-time tracking of the human body[J]. Transaction on Pattern Analysis and Machine Intelligence (S1939-3539), 1996.
[7] Stauffer C, Grimson W E L. Adaptive Background Mixture Models for Real-Time Tracking[C]. Proc Computer Vision & Pattern Recognition, IEEE, 1999.
[8] Elgammal A, Harwood D, Davis L.Non-parametric Model for Background Subtraction[C]. European Conference on Computer Vision. Berlin: Springer, 2000.
[9] Barnich O, Droogenbroeck M V.VIBE: A powerful random technique to estimate the background in video sequences[C]. International Conference on Acoustics, Speech and Signal Processing. Taipei: IEEE, 2009.
[10] Candes E, Li X, Ma Y, et al.Robust principal component analysis[J]. Journal of the ACM (S1557-735X), 2011, 58(3): 11.
[11] Ding X.Bayesian Robust Principal Component Analysis[J]. IEEE Transactions on Image Processing (S1941-0042), 2011, 20(12): 3419.
[12] Jun He.Incremental gradient on the Grassmannian for online foreground and background separation in subsampled video[C]. CVPR. Providence: IEEE, 2012, 157: 1568-1575.
[13] Horn B K P, Schunck B G. Determining Optical Flow[C]. Artificial Intelligence, Washington, 1981.
[14] Lucas B D, Kanade T.An iterative image registration technique with an application to stereo vision[C]. International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann Publishers, 1981.
[15] Bouguet J Y.Pyramidal implementation of the affine Lucas-Kanade feature tracker description of the algorithm[R]. Intel Corporation, 2001.
[16] Dosovitskiy A.Flownet: Learning optical ?ow with convolutional networks[C]. ICCV, 2015.
[17] Ranjan A, Black M J.Optical Flow Estimation using a Spatial Pyramid Network[C]. CVPR, 2016.
[18] Ahmadi A, Patras I.Unsupervised convolutional neural networks for motion estimation[C]. ICIP. Phoenix: IEEE, 2016.
[19] Aubry M.Seeing 3D Chairs: Exemplar Part-Based 2D-3D Alignment Using a Large Dataset of CAD Models[C]. CVPR. Columbus: IEEE, 2014.
[20] Goyette N, Jodoin P M, Porikli F, et al.Changedetection. net: A new change detection benchmark dataset[C]. CVPR. Providence: IEEE, 2012.
[21] Li L.Statistical modeling of complex backgrounds for foreground object detection[J]. IEEE Transactions on Image Processing (S1941-0042), 2004, 13(11): 1459.
[22] Zhou X, Yang C, Yu W.Moving Object Detection by Detecting Contiguous Outliers in the Low-Rank Representation[J]. Transactions on Pattern Analysis and Machine Intelligence (S1939-3539), 2013, 35(3): 597-610.

基于深度卷积神经网络的运动目标检测方法

Moving Object Detection Based on Deep Convolutional Neural Network

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	周培培, 侯幸林. 一种用于图像融合的无监督深度神经网络[J]. 系统仿真学报, 2022, 34(6): 1267-1274.
[2]	曹建芳, 贾一鸣, 闫敏敏, 田晓东. 稳定增强生成对抗网络在壁画的超分辨率重建[J]. 系统仿真学报, 2022, 34(5): 1076-1089.
[3]	冯开团, 袁杰. 基于改进注水算法的离散车间任务分配问题研究[J]. 系统仿真学报, 2022, 34(4): 768-776.
[4]	宁小娟, 李洁茹, 高凡, 王映辉. 基于最佳几何约束和RANSAC的特征匹配算法[J]. 系统仿真学报, 2022, 34(4): 727-734.
[5]	杨敬辉, 刘德康, 杜万和, 邢立宁. 基于图像特征的双目测距系统的研究[J]. 系统仿真学报, 2022, 34(3): 624-632.
[6]	王语童, 马世伟, 杨元睿, 陈超宇. 多机器人组合最大覆盖面积寻优及预测方法[J]. 系统仿真学报, 2022, 34(1): 86-92.
[7]	刘茂山, 纪志成, 王艳, 王建锋. 核块对角表达子空间聚类及收敛性分析[J]. 系统仿真学报, 2021, 33(11): 2533-2544.
[8]	霍志浩, 金炜东, 唐鹏. 融合自监督学习的单帧图像运动视差关键点估计[J]. 系统仿真学报, 2021, 33(11): 2753-2759.
[9]	胡蕾, 王足根, 陈田, 张永梅. 一种改进的SRGAN红外图像超分辨率重建算法[J]. 系统仿真学报, 2021, 33(9): 2109-2118.
[10]	李新利, 邹昌铭, 杨国田, 刘禾. 基于生成式对抗网络的发票图像超分辨率研究[J]. 系统仿真学报, 2021, 33(6): 1307-1314.
[11]	柳有权, 裴雪, 李婉, 刘正雄. 基于邻近目标置信度评估的视觉目标跟踪与定位[J]. 系统仿真学报, 2020, 32(7): 1294-1300.
[12]	汤雯洁, 李若桐, 邓雪松, 司伟鑫, 王琼. ICP优化下的B样条四维CT图像弹性配准方法[J]. 系统仿真学报, 2020, 32(7): 1301-1311.
[13]	王素琴, 张峰, 高宇豆, 石敏. 基于图像序列的学习表情识别[J]. 系统仿真学报, 2020, 32(7): 1322-1330.
[14]	张惠娟, 郭欣琪, 王冬青, 贾金原. 基于DR预测的大规模Web3D场景预加载机制[J]. 系统仿真学报, 2020, 32(7): 1341-1348.
[15]	钱文华, 徐丹, 徐瑾, 何磊, 张波. 东巴画艺术风格绘制[J]. 系统仿真学报, 2020, 32(7): 1349-1359.