曲率滤波-经验模式分解的运动人体目标检测预处理
|
摘要
利用曲率滤波-经验模式分解预处理检测并提取人体目标特征,以降低图像分解运算复杂度,同步增强边缘和纹理特征,提高特征区分性。表现在:(1)在第一层经验模态分解中,以曲率滤波曲面映射原图像的连续平滑曲面,形成包络曲面及均值面,首层分解图像纹理特征显著,以下各层凸显边缘与结构特征;(2)从低分辨率到高分辨率尺度图像匹配出相邻层强边缘区域,易于人体目标轮廓候选区域的提取;(3)以首层分解图纹理特征筛分背景,在相邻层中匹配前景特征区域,形成人体姿态特征的轨迹图,易于判别人体姿态及行为。在人体行为识别实验中,采用曲率滤波-经验模式分解预处理提取的轮廓特征与人体行为典型数据库ground truth对比,在UIUC示例数据的轮廓提取的精度和召回率都达到90%以上。对人体姿态及行为做识别处理,验证了预处理方法的有效性。
Curvature filtering-empirical mode decomposition preprocessing was used to detect and extract human target features. It reduced the computational complexity of image decomposition, enhanced edge and texture features simultaneously, and improved feature differentiation. Its performance was in the following areas:(1) In the first layer of empirical mode decomposition, the continuous smooth surface of the original image was mapped by curvature filtering plane to form the envelope surface and the mean surface. The first layer extracted texture features, and the following layers highlighted structure features.(2) Matching regions had sharp edges from adjacent layers which varied from low-resolution to highresolution, easy to extract the target contour candidate regions.(3) Decomposing the texture features in the first floor to screen the background, matching the foreground feature areas in the adjacent layers to form the trajectory map of the human body posture, it made easy judging the human body posture and behavior. When applying it to human behavior recognition experiments, and comparing it with the ground truth of the human behavior database, the accuracy of the contour extraction and the recall rate of the UIUC sample data were all over 90%. And experiments verify the preprocessing method is effective in human pose recognition and behavior recognition.
Curvature filtering-empirical mode decomposition preprocessing was used to detect and extract human target features. It reduced the computational complexity of image decomposition, enhanced edge and texture features simultaneously, and improved feature differentiation. Its performance was in the following areas:(1) In the first layer of empirical mode decomposition, the continuous smooth surface of the original image was mapped by curvature filtering plane to form the envelope surface and the mean surface. The first layer extracted texture features, and the following layers highlighted structure features.(2) Matching regions had sharp edges from adjacent layers which varied from low-resolution to highresolution, easy to extract the target contour candidate regions.(3) Decomposing the texture features in the first floor to screen the background, matching the foreground feature areas in the adjacent layers to form the trajectory map of the human body posture, it made easy judging the human body posture and behavior. When applying it to human behavior recognition experiments, and comparing it with the ground truth of the human behavior database, the accuracy of the contour extraction and the recall rate of the UIUC sample data were all over 90%. And experiments verify the preprocessing method is effective in human pose recognition and behavior recognition.
引文
[1]Luo Haibo,Xu Lingyun,Hui Bin,et al.Status and prospect of target tracking based on deep learning[J].Infrared and Laser Engineering,2017,46(5):0502002.(in Chinese)
[2]Ren S,He K,Girshick R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2017,39(6):1137-1149.
[3]Wang Xiaofei,Wang Xiaoyi,Shi Xiangyu,et al.Target detection algorithm based on spatial-contextual image one class classification[J].Infrared and Laser Engineering,2015,44(S1):236-240.(in Chinese)
[4]Ke Hongchang,Sun Hongbin.A saliency target area detection method of image sequence[J].Chinese Optics,2015,8(5):768-774.(in Chinese)
[5]Liu Hongbin,Chang Faliang.Moving object detection by optical flow method based on adaptive weight coefficient[J].Optics and Precision Engineering,2016,24(2):460-468.(in Chinese)
[6]Lu Mu,Zhu Ming,Gao Yang,et al.Moving target detection based on dynamic background of cellular automaton[J].Optics and Precision Engineering,2017,25(7):1934-1940.(in Chinese)
[7]Wang Kejia,Ping Ziliang,Sheng Yunlong.Development of image invariant moments-a short overview[J].Chinese Optics Letters,2016,14(9):091001.
[8]You Ruirong,Wang Xinwei,Ren Pengdao,et al.Target observation performance evaluation method for video surveillance based on Johnson criteria[J].Infrared and Laser Engineering,2016,45(12):1217003.(in Chinese)
[9]Subudhi B N,Ghosh S,Nanda P K,et al.Moving object detection using spatio-temporal multilayer compound Markov random field and histogram thresholding based change detection[J].Multimedia Tools and Applications,2017,76(11):13511-13543.
[10]Nunes J C,Bouaoune Y,Delechelle E,et al.Image analysis by bidimensional empirical mode decomposition[J].Image and Vision Computing,2003,21(12):1019-1026.
[11]Bakhtiari S,Agaian S,Mo J.An enhanced empirical mode decomposition based method for image enhancement[C]//IEEE International Conference on Systems,Man,and Cybernetics,2015:2681-2686.
[12]Zhang Shengping,Yao Hongxun,Sun Xin,et al.Action recognition based on overcomplete independent components analysis[J].Information Sciences,2014,281(10):635-647.
[13]Liu Jingen,Benjamin Kuipers,Silvio Savarese.Recognizing human actions by attributes[C]//Computer Vision and Pattern Recognition(CVPR),IEEE Conference on,2011:3337-3344.
[14]Laptev I,Marszalek M,Schmid C,et al.Learning realistic human actions from movies[C]//Proceedings of International Conference on Pattern Recognition,2008:1-8.
[15]Klaser A,Marszalek M,Schmid C.A spatio-temporal descriptor based on 3d gradients[C]//Proceedings of British Machine Vision Association,2008:995-1004.
[16]Liu J,Luo J,Shah M.Recognizing realistic actions from videos"in the wild"[C]//Proceedings of International Conference on Pattern Recognition,2009:1996-2003.
[17]Gong Yuanhao,Sbalzarini Ivo F.Curvature filters efficiently reduce certain variational energies[J].IEEE Transaction on Image Processing,2017,26(4):1786-1798.
[18]Brito-Loeza C,Chen K,Uc-Cetina V.Image denoising using the Gaussian curvature of the image surface[J].Numerical Methods for Partial Differential Equations,2016,32(3):1066-1089.
[19]Martins P,Carvalho P,Gatta C.On the completeness of feature-driven maximally stable extremal regions[J].Pattern Recognition Letters,2016,74:9-16.
[2]Ren S,He K,Girshick R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2017,39(6):1137-1149.
[3]Wang Xiaofei,Wang Xiaoyi,Shi Xiangyu,et al.Target detection algorithm based on spatial-contextual image one class classification[J].Infrared and Laser Engineering,2015,44(S1):236-240.(in Chinese)
[4]Ke Hongchang,Sun Hongbin.A saliency target area detection method of image sequence[J].Chinese Optics,2015,8(5):768-774.(in Chinese)
[5]Liu Hongbin,Chang Faliang.Moving object detection by optical flow method based on adaptive weight coefficient[J].Optics and Precision Engineering,2016,24(2):460-468.(in Chinese)
[6]Lu Mu,Zhu Ming,Gao Yang,et al.Moving target detection based on dynamic background of cellular automaton[J].Optics and Precision Engineering,2017,25(7):1934-1940.(in Chinese)
[7]Wang Kejia,Ping Ziliang,Sheng Yunlong.Development of image invariant moments-a short overview[J].Chinese Optics Letters,2016,14(9):091001.
[8]You Ruirong,Wang Xinwei,Ren Pengdao,et al.Target observation performance evaluation method for video surveillance based on Johnson criteria[J].Infrared and Laser Engineering,2016,45(12):1217003.(in Chinese)
[9]Subudhi B N,Ghosh S,Nanda P K,et al.Moving object detection using spatio-temporal multilayer compound Markov random field and histogram thresholding based change detection[J].Multimedia Tools and Applications,2017,76(11):13511-13543.
[10]Nunes J C,Bouaoune Y,Delechelle E,et al.Image analysis by bidimensional empirical mode decomposition[J].Image and Vision Computing,2003,21(12):1019-1026.
[11]Bakhtiari S,Agaian S,Mo J.An enhanced empirical mode decomposition based method for image enhancement[C]//IEEE International Conference on Systems,Man,and Cybernetics,2015:2681-2686.
[12]Zhang Shengping,Yao Hongxun,Sun Xin,et al.Action recognition based on overcomplete independent components analysis[J].Information Sciences,2014,281(10):635-647.
[13]Liu Jingen,Benjamin Kuipers,Silvio Savarese.Recognizing human actions by attributes[C]//Computer Vision and Pattern Recognition(CVPR),IEEE Conference on,2011:3337-3344.
[14]Laptev I,Marszalek M,Schmid C,et al.Learning realistic human actions from movies[C]//Proceedings of International Conference on Pattern Recognition,2008:1-8.
[15]Klaser A,Marszalek M,Schmid C.A spatio-temporal descriptor based on 3d gradients[C]//Proceedings of British Machine Vision Association,2008:995-1004.
[16]Liu J,Luo J,Shah M.Recognizing realistic actions from videos"in the wild"[C]//Proceedings of International Conference on Pattern Recognition,2009:1996-2003.
[17]Gong Yuanhao,Sbalzarini Ivo F.Curvature filters efficiently reduce certain variational energies[J].IEEE Transaction on Image Processing,2017,26(4):1786-1798.
[18]Brito-Loeza C,Chen K,Uc-Cetina V.Image denoising using the Gaussian curvature of the image surface[J].Numerical Methods for Partial Differential Equations,2016,32(3):1066-1089.
[19]Martins P,Carvalho P,Gatta C.On the completeness of feature-driven maximally stable extremal regions[J].Pattern Recognition Letters,2016,74:9-16.