期刊文献+

背景差分与三帧差分结合的运动目标检测算法 被引量:45

A Method for Moving Object Detection Based on Background Subtraction and Three- Frame Differencing
下载PDF
导出
摘要 针对基于混合高斯模型背景差分法对光照突变敏感的问题,提出了背景差分法与三帧间差分法相结合的运动目标检测算法;首先利用当前帧与混合高斯模型建立的背景模型差分,快速检测出运动变化区域;然后,通过与设定的阈值比较,判断场景内是否发生光照突变,场景中若未发生光照突变,则采用混合高斯模型背景差分法提取运动目标,若发生光照突变,采用三帧间差分法提取运动目标;实验结果表明,在光线发生突变情况下,文中提出的算法同样能够取得较好的检测效果,具有很强的适应性和鲁棒性,可用于实时监测系统。 The background subtraction method based on Gaussian mixture model is sensitive to light mutation,a combining method of background subtraction and three-frame differencing is proposed to make up its deficiency.First,the current image minuses the background model to detect the moving area rapidly,then compared to the threshold value set previously to judge whether light mutation happens in the monitoring area.If there is no light mutation,the method of the background subtraction based on Gaussian mixture model is selected to extract moving objects;in the situation of light mutation,the method of the three-frame differencing is selected to extract moving objects.At last,the experiment shows that the proposed method can also have a better detection result when light mutation happens.Meanwhile the robustness and suitability are also acceptable,and it can be applied into real-time video monitoring system.
出处 《计算机测量与控制》 北大核心 2013年第12期3315-3318,共4页 Computer Measurement &Control
基金 高等学校博士学科点专项科研基金(20113227110007)
关键词 光照突变 混合高斯模型 背景差分法 三帧差分法 实时监测系统 light mutation Gaussian mixture model method of background subtraction method of three-frame differencing real-time video monitoring system
  • 相关文献

参考文献12

  • 1刘静,王玲.混合高斯模型背景法的一种改进算法[J].计算机工程与应用,2010,46(13):168-170. 被引量:55
  • 2OHTA N. A. Statistical approach to background subtraction for surveillance systems EA]. In pro. Of international Conference Computer Vision -C]. 2001:481 - 486.
  • 3姚会,苏松志,王丽,李绍滋.基于改进的混合高斯模型的运动目标检测方法[J].厦门大学学报(自然科学版),2008,47(4):505-510. 被引量:22
  • 4Long W, Yang Y. Stationary background generation. An alterna- tive to the difference of two images [J]. Pattern Recognition, 1990, 23 (12): 1351-1359.
  • 5FRI[DMAN N, RUSSELL S. Image segmentation in video se- quences A probabilistic approach [-A], In: Proc. Of the 13th Con- ference on Uncertainty in Artificial Intelligence i-C]. San Francisco, 1997: 175-181.
  • 6STAUFFER C, GRIMSON W E L. Adaptive background mixture models for real-time tracking f-A. In: Proc. The IEEE Computer Society Conference on Computer Vision and Pattern Recognition [C]. 1999, 2: 246-252.
  • 7STAUFFER C. , GRIMSON W E L. Learning patterns of activity using real-time tracking [JT. IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 2000, 22 (8) : 747 - 757.
  • 8CHALIDABHONGSE T H, KIM K, HARWOOD D, et al. A per- turbation method for evaluating background subtraction algorithms [-A. Nice, France: Proceeding of the Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance [-C. 2003 : 11 - ]2.
  • 9Wang L, Wen M, et al. Background subtraction using incremental subspace learning EA-]. San Antonio, USA: International Confer- ence on Image Processing i-C]. 2007:45 - 48.
  • 10吕国亮,赵曙光,赵俊.基于三帧差分和连通性检验的图像运动目标检测新方法[J].液晶与显示,2007,22(1):87-93. 被引量:36

二级参考文献43

共引文献133

同被引文献298

引证文献45

二级引证文献216

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部