《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (2): 514-520.DOI: 10.11772/j.issn.1001-9081.2021122112

• 多媒体计算与计算机仿真 • 上一篇    

基于可靠性低秩因子分解和泛化差异性差分的运动目标检测

汪鹏1,2,3, 张大蔚1,2,3, 陆正军4(), 李林昊1,2,3   

  1. 1.河北工业大学 人工智能与数据科学学院, 天津 300401
    2.河北省大数据计算重点实验室(河北工业大学), 天津 300401
    3.河北省数据驱动工业智能工程研究中心(河北工业大学), 天津 300401
    4.军事科学院 国防工程研究院, 北京 100036
  • 收稿日期:2021-12-18 修回日期:2022-05-13 接受日期:2022-05-18 发布日期:2022-06-30 出版日期:2023-02-10
  • 通讯作者: 陆正军
  • 作者简介:汪鹏(1978—),男,河北邯郸人,副教授,博士,主要研究方向:图像处理
    张大蔚(1997—),男,河北滦县人,硕士研究生,主要研究方向:计算机视觉、机器学习
    李林昊(1989—),男,山东威海人,助理教授,博士,CCF会员,主要研究方向:量化和哈希学习、稀疏信号恢复、背景建模和前景检测。
  • 基金资助:
    国家重点研发计划项目(2019YFC1904601);国家自然科学基金资助项目(61902106);河北省高等学校科学技术研究项目(QN2021213)

Moving object detection based on reliability low-rank factorization and generalized diversity difference

Peng WANG1,2,3, Dawei ZHANG1,2,3, Zhengjun LU4(), Linhao LI1,2,3   

  1. 1.School of Artificial Intelligence,Hebei University of Technology,Tianjin 300401,China
    2.Hebei Province Key Laboratory of Big Data Calculation (Hebei University of Technology),Tianjin 300401,China
    3.Hebei Data Driven Industrial Intelligent Engineering Research Center (Hebei University of Technology),Tianjin 300401,China
    4.Defense Engineering Institute,Academy of Military Sciences,Beijing 100036,China
  • Received:2021-12-18 Revised:2022-05-13 Accepted:2022-05-18 Online:2022-06-30 Published:2023-02-10
  • Contact: Zhengjun LU
  • About author:WANG Peng, born in 1978, Ph. D., associate professor. His research interests include image processing.
    ZHANG Dawei, born in 1997, M. S. candidate. His research interests include computer vision, machine learning.
    LI Linhao, born in 1989, Ph. D., assistant professor. His research interests include quantization and hash learning, sparse signal recovery, background modeling and foreground detection.
  • Supported by:
    National Key Research and Development Program of China(2019YFC1904601);National Natural Science Foundation of China(61902106);Science and Technology Research Project of Higher Education Institutes in Hebei Province(QN2021213)

摘要:

运动目标检测旨在分离视频的背景与前景,然而常用的低秩因子分解法往往难以综合地处理动态背景和间歇性运动的问题。考虑到背景减除后的偏态噪声分布具有潜在的背景修正作用,提出一种基于可靠性低秩因子分解和泛化差异性差分的运动目标检测模型。首先,利用时间维度像素分布的峰值位置以及偏态分布性质选取一个不含离群像素的子序列,并计算该子序列的中值以形成静态背景;其次,利用非对称拉普拉斯分布对静态背景减除后的噪声建模,并把基于空间平滑的建模结果作为可靠性权重参与到低秩因子分解中,以此建模综合背景(含有动态背景);最后,依次利用时间和空间连续约束提取前景。其中,针对时间连续性,提出了泛化差异性差分约束,从而通过相邻视频帧的差异信息抑制前景边缘的扩增。实验结果表明,与PCP、DECOLOR、LSD、TVRPCA、E-LSD、GSTO六种模型相比,所提模型的F-measure值最高。由此可知,所提模型在动态背景、间歇性运动等复杂场景中能有效提高前景的检测精度。

关键词: 非对称噪声建模, 低秩因子分解, 中值背景建模, 运动目标检测

Abstract:

Moving object detection aims to separate the background and foreground of the video, however, the commonly used low-rank factorization methods are often difficult to comprehensively deal with the problems of dynamic background and intermittent motion. Considering that the skewed noise distribution after background subtraction has potential background correction effect, a moving object detection model based on the reliability low-rank factorization and generalized diversity difference was proposed. There were three steps in the model. Firstly, the peak position and the nature of skewed distribution of the pixel distribution in the time dimension were used to select a sub-sequence without outlier pixels, and the median of this sub-sequence was calculated to form the static background. Secondly, the noise after static background subtraction was modeled by asymmetric Laplace distribution, and the modeling results based on spatial smoothing were used as reliability weights to participate in low-rank factorization to model comprehensive background (including dynamic background). Finally, the temporal and spatial continuous constraints were adopted in proper order to extract the foreground. Among them, for the temporal continuity, the generalized diversity difference constraint was proposed, and the expansion of the foreground edge was suppressed by the difference information of adjacent video frames. Experimental results show that, compared with six models such as PCP(Principal Component Pursuit), DECOLOR(DEtecting Contiguous Outliers in the Low-Rank Representation), LSD(Low-rank and structured Sparse Decomposition), TVRPCA(Total Variation regularized Robust Principal Component Analysis), E-LSD(Extended LSD) and GSTO(Generalized Shrinkage Thresholding Operator), the proposed model has the highest F-measure. It can be seen that this model can effectively improve the detection accuracy of foreground in complex scenes such as dynamic background and intermittent motion.

Key words: asymmetric noise modeling, low-rank factorization, median background modeling, moving object detection

中图分类号: