基于混合高斯模型的运动目标检测
|
摘要
针对运动目标检测中的背景复杂度高、视频数据计算量大等问题,且为避免计算不同复杂程度的视频背景,并能够准确地获取所需要的运动目标,提出了一种基于混合高斯模型的运动目标检测方法.首先采用混合高斯模型获取运动目标特征;然后利用中值滤波方法去除视频目标运动特征中的背景噪声;最后依据形态学运算方法对通过统计直方图得到的运动显著图进行处理,从而获取最终的运动目标.对标准视频序列集的检测表明,利用该算法获取的运动目标不仅能抑制背景噪声,而且精准度和误差都优于普通的视频运动目标检测算法.
Aiming at tackling the problems of high background complexity and large amount of video data in moving target detection,and in order to avoid computing the background of different complexity,a moving target detection method based on Gaussian mixture model is proposed.First,a Gaussian mixture model is used to acquire the features of the moving target,and then median filtering is applied to remove the background noise in the motion feature of the video target.Finally,by taking advantage of the morphological operation method,the motion image obtained by the statistical histogram is processed to obtain the final moving target.Tested on the standard video sequence set,the moving target obtained using the proposed algorithm can not only suppress the background noise,but also be better than the ordinary video moving target detection algorithm both in the accuracy improvement and error attenuation.
Aiming at tackling the problems of high background complexity and large amount of video data in moving target detection,and in order to avoid computing the background of different complexity,a moving target detection method based on Gaussian mixture model is proposed.First,a Gaussian mixture model is used to acquire the features of the moving target,and then median filtering is applied to remove the background noise in the motion feature of the video target.Finally,by taking advantage of the morphological operation method,the motion image obtained by the statistical histogram is processed to obtain the final moving target.Tested on the standard video sequence set,the moving target obtained using the proposed algorithm can not only suppress the background noise,but also be better than the ordinary video moving target detection algorithm both in the accuracy improvement and error attenuation.
引文
[1]陆军,李凤玲,姜迈.摄像机运动下的动态目标检测与跟踪[J].哈尔滨工程大学学报,2008,29(8):831-835.
[2]熊卫华,向磊,李俊峰,等.背景减除与帧间差分相结合的运动目标检测方法[EB/OL].[2018-10-12].http://cpfd.cnki.com.cn/Article/CPFDTOTAL-KZLL201107001151.htm.
[3]谢之宇,蒋晓瑜,汪熙,等.改进的帧差法在目标匹配中的应用[J].计算机工程与应用,2011,47(34):208-211.
[4]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.
[5]Xia Y,Shi X,Song G,et al.Towards improving quality of video-based vehicle counting method for traffic flow estimation[J].Signal Processing,2016,120:672-681.
[6]桂本烨,钱徽,朱淼良.一种优化梯度计算的改进HS光流算法[J].中国图象图形学报,2005,10(8):1052-1058.
[7]Meyer D,Denzler J,Niemann H.Model based extraction of articulated objects in image sequences for gait analysis[C]//Proceedings of IEEE International Conference on Image Processing,Santa Barbara,1997:78-81.
[8]汪巍.动态场景下运动目标检测与跟踪算法的研究[D].合肥:合肥工业大学,2012.
[9]陈栋喜.基于改进自适应混合高斯模型的背景建模及对象计数方法研究[D].南京:南京航空航天大学,2013.
[10]黄东军,杨颖华.基于改进混合高斯模型的运动物体检测研究[J].计算机应用研究,2017,34(6):1862-1866.
[11]孙炜烨.动态场景下运动目标检测算法的研究[D].天津:河北工业大学,2012.
[12]高智勇,唐文峰,贺良杰.基于运动显著性的移动镜头下的运动目标检测[J].计算机应用,2016,36(6):1692-1698.
[13]李晓悦.基于改进CV模型的工业CT图像分割算法研究[D].大连:大连理工大学,2014.
[14]Cheng M M,Zhang G X,Mitra N J,et al.Global contrast based salient region detection[EB/OL].[2018-10-12].http://vecg.cs.ucl.ac.uk/Projects/Smart Geometry/contrast_sal iency/paper_docs/contrast_saliency_small_cvpr11.pdf.
[15]高秀丽.融合深度信息的视觉注意模型研究及其应用[D].杭州:杭州电子科技大学,2013.
[16]胡燕,王慧琴,马宗方.改进混合高斯模型的自适应烟雾图像分割算法[J].计算机辅助设计与图形学学报,2016,28(7):1138-1145.
[17]董俊宁,杨词慧.空间约束混合高斯运动目标检测[J].中国图象图形学报,2016,21(5):588-594.
[18]Choi Y J,Lee J S,Cho W D.Robust extraction of moving objects based on hue and hue gradient[C]//Universal Access in Human-computer Interaction,Ambient Interaction.Berlin:Springer,2007:784-791.
[19]郭伟,高媛媛,刘鑫焱.改进的基于混合高斯模型的运动目标检测算法[J].计算机工程与应用,2016,52(13):195-200.
[2]熊卫华,向磊,李俊峰,等.背景减除与帧间差分相结合的运动目标检测方法[EB/OL].[2018-10-12].http://cpfd.cnki.com.cn/Article/CPFDTOTAL-KZLL201107001151.htm.
[3]谢之宇,蒋晓瑜,汪熙,等.改进的帧差法在目标匹配中的应用[J].计算机工程与应用,2011,47(34):208-211.
[4]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.
[5]Xia Y,Shi X,Song G,et al.Towards improving quality of video-based vehicle counting method for traffic flow estimation[J].Signal Processing,2016,120:672-681.
[6]桂本烨,钱徽,朱淼良.一种优化梯度计算的改进HS光流算法[J].中国图象图形学报,2005,10(8):1052-1058.
[7]Meyer D,Denzler J,Niemann H.Model based extraction of articulated objects in image sequences for gait analysis[C]//Proceedings of IEEE International Conference on Image Processing,Santa Barbara,1997:78-81.
[8]汪巍.动态场景下运动目标检测与跟踪算法的研究[D].合肥:合肥工业大学,2012.
[9]陈栋喜.基于改进自适应混合高斯模型的背景建模及对象计数方法研究[D].南京:南京航空航天大学,2013.
[10]黄东军,杨颖华.基于改进混合高斯模型的运动物体检测研究[J].计算机应用研究,2017,34(6):1862-1866.
[11]孙炜烨.动态场景下运动目标检测算法的研究[D].天津:河北工业大学,2012.
[12]高智勇,唐文峰,贺良杰.基于运动显著性的移动镜头下的运动目标检测[J].计算机应用,2016,36(6):1692-1698.
[13]李晓悦.基于改进CV模型的工业CT图像分割算法研究[D].大连:大连理工大学,2014.
[14]Cheng M M,Zhang G X,Mitra N J,et al.Global contrast based salient region detection[EB/OL].[2018-10-12].http://vecg.cs.ucl.ac.uk/Projects/Smart Geometry/contrast_sal iency/paper_docs/contrast_saliency_small_cvpr11.pdf.
[15]高秀丽.融合深度信息的视觉注意模型研究及其应用[D].杭州:杭州电子科技大学,2013.
[16]胡燕,王慧琴,马宗方.改进混合高斯模型的自适应烟雾图像分割算法[J].计算机辅助设计与图形学学报,2016,28(7):1138-1145.
[17]董俊宁,杨词慧.空间约束混合高斯运动目标检测[J].中国图象图形学报,2016,21(5):588-594.
[18]Choi Y J,Lee J S,Cho W D.Robust extraction of moving objects based on hue and hue gradient[C]//Universal Access in Human-computer Interaction,Ambient Interaction.Berlin:Springer,2007:784-791.
[19]郭伟,高媛媛,刘鑫焱.改进的基于混合高斯模型的运动目标检测算法[J].计算机工程与应用,2016,52(13):195-200.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |