动态特征块匹配的背景更新在运动检测的应用
DOI:
作者:
作者单位:

长春理工大学光电工程学院长春130022

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:

吉林省科技发展计划(20160520018JH)、国家自然科学基金(61605016)项目资助


Background updating based on dynamic feature block matching for the motion detection
Author:
Affiliation:

School of Photoelectric Engineering, Changchun University of Science and Technology, Changchun 130022,China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为解决传统监控设备视场小、非智能等缺陷,结合全景成像技术和计算机视觉技术,建立无人环境下外来入侵自动检测系统,从而实现了全景监控视场下运动目标快速准确的检测及跟踪。该技术关键在于如何在复杂的动态背景下有效地提取运动目标,为此提出一种基于动态特征块匹配的自适应背景更新算法。在采用帧间差分与背景差分融合算法检测到目标的基础上,利用目标的矩信息进行跟踪,避免了全景视觉下颜色及轮廓特征缺失的弊端。根据目标的轮廓及位置提取特征块,将视频序列的每一帧图像与初始背景图像进行特征块区域的局部匹配,首先通过分析特征块图像的颜色特征,构建基于区间统计的RGB颜色直方图,提取颜色特征序列。然后通过计算序列相关性来判断该区域是否需要背景更新,从而降低对单个像素更新的冗余计算。实验表明,该更新算法具有较强的鲁棒性和可行性,能够有效提高监控系统的稳定性。

    Abstract:

    To overcome the limitations of small fieldview and nonintelligent of traditional monitoring systems, this work combines panoramic imaging and computer vision technology to establish an automatic intrusion detection system in unmanned environment. The proposed system can achieve fast and accurate moving object detection and tracking in panoramic monitoring field. To solve the key challenge on effectively extracting moving objects in complex and dynamic background, this paper presents an adaptive background updating algorithm based on dynamic matching between feature blocks. On the basis of target detection using the fusion algorithm with frame differencing background subtraction, the torque information is utilized for tracking the target so as to avoid the lack of color and contour features under the panoramic view. The feature block si then extracted according to the outline and position of the target, in which the local matching of feature area is conducted between video sequences of each frame image and the initial background image. The color feature of the feature block is firstly analyzed, and a RGB color histogram is established based on interval statistics. Thus, the colour feature sequences are obtained.wWhether the region background updates is then determined by calculating the correlation between the two sequences to reducethe computation of a single pixel to be updated. Experiments show that the proposed algorithm is robust and feasible, and it can effectively improve the stability of the monitoring system.

    参考文献
    相似文献
    引证文献
引用本文

李艳荻,徐熙平,陈江,王鹤程.动态特征块匹配的背景更新在运动检测的应用[J].仪器仪表学报,2017,38(2):445-453

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2017-07-20
  • 出版日期: