A Method for Detecting Moving Objects and Eliminating Shadow in Natural Scene

doi:10.7652/xjtuxb201312005

您当前的位置：

首页 >

文章列表页 >

A Method for Detecting Moving Objects and Eliminating Shadow in Natural Scene

更新时间：2025-07-09

- A Method for Detecting Moving Objects and Eliminating Shadow in Natural Scene
- Vol. 47, Issue 12, Pages: 26-31(2013)
- 作者机构：
  
  武汉大学电子信息学院,武汉,430072
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb201312005
  CLC： TP391
- Online First：10 December 2013，
  
  Published：2013
- 稿件说明：
移动端阅览
A Method for Detecting Moving Objects and Eliminating Shadow in Natural Scene[J]. 2013, 47(12): 26-31.
DOI：

A Method for Detecting Moving Objects and Eliminating Shadow in Natural Scene[J]. 2013, 47(12): 26-31. DOI： 10.7652/xjtuxb201312005.

摘要

针对自然场景下运动目标检测中投射阴影使前景掩模检测精度低的问题

提出一种基于双阈值多分辨率的运动目标检测和多属性阴影剔除方法(MRPS)。首先根据不同分辨率下点的颜色属性

用双阈值混合高斯模型来检测运动掩模

用计算颜色模型来克服光照变化和轻度阴影的影响; 然后根据阴影的光学属性

确定运动掩模中的潜在阴影区域

根据该区域外轮廓的位置属性对其分类; 最后通过分析边缘属性恢复与阴影相连的目标背光区域

保留运动掩模中真实的运动目标区域。用公共测试图像的shadow序列对该方法进行了验证

并与狄利克雷过程(DPGMM)方法进行了比较

结果表明

该方法获取的前景掩模能更加有效地克服投射阴影的影响

前景检测精度提高了7.08%。

Abstract

A method for detecting moving objects based on dual-threshold and multi-resolution Gaussian mixture models and for eliminating shadow based on multi-attribute(named MRPS)is proposed to address the problem of low foreground mask detection accuracy caused by cast shadows under natural scene in detecting moving objects. First

the MRPS detects moving motion mask using the dual-threshold and multi-resolution Gaussian mixture models and overcomes the influence of illumination change and light shadows using the computational color model according to the point's color properties under different resolutions. Then

the potential shadow areas are determined in the motion mask from the optical properties

and the properties of the potential shadow areas are classified through analyzing the location properties for their outer contours. Finally

the backlighting object areas connecting with the shadows is recovered and the real object areas in the moving mask are retained by analyzing edge properties. The proposed algorithm is verified using the shadow sequences in the public test image sequences and compared with Dirichlet processes Gaussian mixture models(DPGMM)method. Experimental results show that the foreground obtained by the MPRS overcomes the influence of cast shadows more effectively

and the foreground detection precision increases by 7.08%.

关键词

Keywords

references

CHEUNG S, KAMATH C. Robust background subtraction with foreground validation for urban traffic video [J]. Eurasip Journal on Applied Signal Processing, 2005, 2005(14): 2330-2340.

BRAHME Y B, KULKARNI P S. An implementation of moving object detection, tracking and counting objects for traffic surveillance system [C]∥Proceedings of International Conference on Computational Intelligence and Communication Systems. Piscataway, NJ, USA: IEEE, 2011: 143-148.

WEI L, NGUYEN A V, LEE E J. Real-time recognition of humans by their walk [J]. International Journal of Intelligent Information and Database Systems, 2011, 5(1): 24-38.

GOYETTE N, JODOIN P M, PORIKLI F, et al. Changedetection. net: a new change detection benchmark dataset [C]∥Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Piscataway, NJ, USA: IEEE, 2012: 1-8.

1st IEEE Change Detection Workshop. Change detection net video database [EB/OL]. [2013-01-01]. http:∥wordpress-jodoin.dmi.usherb.ca/static/dataset/dataset.zip.

HOFMANN M, TIEFENBACHER P, RIGOLL G. Background segmentation with feedback: the pixel-based adaptive segmenter [C]∥Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Piscataway, NJ, USA: IEEE, 2012: 38-43.

HERAS E R, SIKORA T. Complementary background models for the detection of static and moving objects in crowded environments [C]∥Proceedings of IEEE International Conference on Advanced Video and Signal Based Surveillance. Piscataway, NJ, USA: IEEE, 2011: 71-76.

ZIVKOVIC Z. Improved adaptive Gaussian mixture model for background subtraction [C]∥Proceedings of International Conference on Pattern Recognition. Piscataway, NJ, USA: IEEE, 2004: 28-31.

SCHICK A, BAUML M, STIEFELHAGEN R. Improving foreground segmentations with probabilistic superpixel Markov random fields [C]∥Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Piscataway, NJ, USA: IEEE, 2012: 27-31.

HAINES T S F, XIANG Tao. Background subtraction with Dirichlet processes [C]∥Proceedings of 12th European Conference on Computer Vision. Berlin, Germany: Springer Verlag, 2012: 99-113.

STAUFFER C, GRIMSON W E L. Adaptive background mixture models for real-time tracking [C]∥Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Los Alamitos, CA, USA: IEEE, 1999: 246-252.

屠礼芬, 仲思东, 彭祺. 基于序列图像的运动目标检测 [J]. 应用光学, 2012, 33(5): 899-903.

TU Lifen, ZHONG Sidong, PENG Qi. Moving object detection based on image sequences [J]. Journal of Applied Optics, 2012, 33(5): 899-903.

DONG Xia, WANG Kedian, JIA Guohua. Moving object and shadow detection based on RGB color space and edge ratio [C]∥Proceedings of the 2009 2nd International Congress on Image and Signal Processing. Piscataway, NJ, USA: IEEE, 2009: 5301770-1-5301770-5.

马义德, 朱望飞, 安世霞, 等. 改进的基于高斯混合模型的运动目标检测方法 [J]. 计算机应用, 2007, 27(10): 2545-2548.

MA Yide, ZHU Wangfei, AN Shixia, et al. Improved moving objects detection method based on Gaussian mixture model [J]. Computer Applications, 2007, 27(10): 2545-2548.

BRADSKI G, KAEBLER A. 学习OpenCV [M]. 于仕琪, 刘瑞祯, 编译. 北京: 清华大学出版社, 2009: 150-155.

SALVADOR E, CAVALLARO A, EBRAHIMI T. Cast shadow segmentation using invariant color features [J]. Computer Vision and Image Understanding, 2004, 95(2): 238-259.[17] CANNY J. A computational approach to edge detection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6): 679-698.

陈炳文, 王文伟, 杨文英. 基于图像块和边缘增强的运动目标检测 [J]. 计算机工程, 2010, 36(17): 192-194.

CHEN Bingwen, WANG Wenwei, YANG Wenying. Moving object detection based on image block and edge enhancement [J]. Computer Engineering, 2010, 36(17): 192-194.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

An Iris Recognition Method Using Log-Gabor Filter and Multi-Resolution Analysis

An Adaptive Shadow Detection Algorithm Using Edge Texture and Sampling Deduction

Study on the Thermodynamic Properties in the Anomalous Region of Supercritical Water

Effects of Expansion Angle Crater Film Hole on the Film Cooling Characteristics of a Double-Wall Structure

Knowledge Graph Representation Learning Driven by Semantic Discrimination and Subgraph Integration

Related Author

唐荣年

韩九强

张新曼

姜柯

李艾华

苏延召

Related Institution

School of Electronics and Information Engineering, Xi'an Jiaotong University

502 Faculty, The Second Artillery Engineering University

⁰