基于自适应Lab色差阈值的ViBe运动目标检测算法
|
摘要
针对视觉背景提取算法(ViBe)对光照变化和运动阴影敏感、提取的运动区域容易产生空洞的问题,本文提出了基于自适应Lab色差阈值的ViBe运动目标检测算法。根据图像的局部背景亮度与色彩的空间频率对人眼视觉的影响,自适应的确定每个像素点的色差阈值,用于像素点与背景模型的匹配;然后,利用邻域像素点的空间一致性原则,对检测结果进行修正;最后,统计各连通域的面积,去除小面积的运动目标。实验结果表明,本算法可以有效的适应光照变化、抑制运动阴影、填补运动区域的空洞,具有比ViBe算法更好的检测效果。
Visual Background Extractor(ViBe) is a popular moving object detection algorithm,which has a high efficiency and a good detection effect.However,ViBe algorithm is sensitive to illumination changes and motion shadows and extracts movement area incompletely.At present,most researchers improve ViBe algorithm in gray space,which does not effectively use the color information of the image.Related research shows that the moving object detection algorithm based on Lab color space has higher recognition rate and lower false detection rate.Therefore,this paper studies ViBe algorithm in Lab color space and proposes a ViBe moving object detection algorithm based on adaptive Lab color difference threshold.For matching pixels with background model,the algorithm adaptively determines color difference threshold of each pixel according to the influence of local changes in luminance and color spatial frequency to human visual system.Then,the algorithm makes use of space consistency of neighborhood pixels to correct the result of pixel classification.Finally,it counts pixel numbers of all connecting areas to remove small moving objects.Experiments show that the algorithm proposed is less sensitive to illumination changes and motion shadows,and makes extracted movement area more complete than the original ViBe algorithm both in indoor and outdoor environments.
Visual Background Extractor(ViBe) is a popular moving object detection algorithm,which has a high efficiency and a good detection effect.However,ViBe algorithm is sensitive to illumination changes and motion shadows and extracts movement area incompletely.At present,most researchers improve ViBe algorithm in gray space,which does not effectively use the color information of the image.Related research shows that the moving object detection algorithm based on Lab color space has higher recognition rate and lower false detection rate.Therefore,this paper studies ViBe algorithm in Lab color space and proposes a ViBe moving object detection algorithm based on adaptive Lab color difference threshold.For matching pixels with background model,the algorithm adaptively determines color difference threshold of each pixel according to the influence of local changes in luminance and color spatial frequency to human visual system.Then,the algorithm makes use of space consistency of neighborhood pixels to correct the result of pixel classification.Finally,it counts pixel numbers of all connecting areas to remove small moving objects.Experiments show that the algorithm proposed is less sensitive to illumination changes and motion shadows,and makes extracted movement area more complete than the original ViBe algorithm both in indoor and outdoor environments.
引文
[1] Yazdi M,Bouwmans T.New trends on moving object detection in video images captured by a moving camera:A survey[J].Computer Science Review,2018,28:157-177.
[2] AN Yun-feng,PAN Hai-peng.A new method for moving bubbles detection based on frame difference[J].Journal of Optoelectronics·Laser,2016,27(9):963-972.安云凤,潘海鹏.一种新的三帧差运动气泡检测方法[J].光电子·激光,2016,27(9):963-972.
[3] BAI Jun-qi.Infrared target detection method based on OTSU[J].Command Information System and Technology,2017,8(2):52-55.白俊奇.基于最大类间方差法的红外目标检测方法[J].指挥信息系统与技术,2017,8(2):52-55.
[4] Goyal K,Singhai J.Review of background subtraction methods using Gaussian mixture model for video surveillance systems[J].Artificial Intelligence Review,2017,(3):1-19.
[5] Liu W,Yu H,Yuan H,et al.Effective background modelling and subtraction approach for moving object detection[J].IET Computer Vision,2015,9(1):13-24.
[6] Barnich O,Droogenbroeck M.V.ViBe:A universal background subtraction algorithm for video sequences[J].IEEE Transactions on Image Processing,2011,20(6):1709-1724.
[7] YU Ye,CAO Ming-wei,YUE Feng.EVibe:An improved Vibe algorithm for detecting moving objects[J].Chinese Journal of Scientific Instrument,2014,35(4):924-931.余烨,曹明伟,岳峰.EVibe:一种改进的Vibe运动目标检测算法[J].仪器仪表学报,2014,35(4):924-931.
[8] CHEN Xing-ming,LIAO Juan,LI Bo,et al.Foreground detection based on modified ViBe in dynamic background[J].Optics and Precision Engineering,2014,22(9):2545-2552.陈星明,廖娟,李勃,等.动态背景下基于改进视觉背景提取的前景检测[J].光学精密工程,2014,22(9):2545-2552.
[9] MO Shao-wen,DENG Xin-pu,WANG Shuai,et al.Moving object detection algorithm based on improved visual background extractor[J].Acta Optica Sinica,2016,36(6):0615001.莫邵文,邓新蒲,王帅,等.基于改进视觉背景提取的运动目标检测算法[J].光学学报,2016,36(6):0615001.
[10] MIN Wen-dong,GUO Xiao-guang,HAN Qing.An improved ViBe algorithm and its application in traffic video processing[J].Optics and Precision Engineering,2017,25(3):806-811.闵卫东,郭晓光,韩清.改进的ViBe算法及其在交通视频处理中的应用[J].光学精密工程,2017,25(3):806-811.
[11] LIU Yong,XIA Dao-pin,QIU Qin-jun,et al.Research on visual background extractor to identify the vehicles based on edge similarity[J].International Journal of Wireless and Mobile Computing,2017,13(1):8-14.
[12] LUO Yuan,ZHOU Han-xing,TAN Qin,et al.Key frame extraction of surveillance video based on moving object detection and image similarity[J].Pattern Recognition and Image Analysis,2018,28(2):225-231.
[13] MIN Wei-dong,FAN Meng-dan,GUO Xiao-guang,et al.A new approach to track multiple vehicles with the combination of robust detection and two classifiers[J].IEEE Transactions on Intelligent Transportation Systems,2018,19(1):174-186.
[14] YANG Shi-yu,HAO Kuang-rong,DING Yong-sheng,et al.Improved visual background extractor with adaptive range change[J].Memetic Computing,2018,10(1):53-61.
[15] DING Ying,QIAN Feng,FAN Jing-tao,et al.Study on moving object detection algorithm based on different color space[J].Journal of Changchun University of Science and Technology (Natural Science Edition),2012,35(4):1-4.丁莹,钱锋,范静涛,等.基于不同颜色空间的运动目标检测算法分析[J].长春理工大学学报( 自然科学版),2012,35(4):1-4.
[16] WANG Ke,LU Chang-de,YUE Wan-de,et al.Color harmony system based on lab perceptual uniform color space[J].Journal of Northwestern Polytechnical University,2004,22(6):695-698.王可,陆长德,乐万德,等.基于Lab均匀色彩空间的色彩调和系统[J].西北工业大学学报,2004,22(6):695-698.
[17] YANG Xian,LI Bo,DING Wen,et al.Local adaptive perceptual color difference for color image edge detection[J].Journal of Electronics & Information Technology,2012,34(9):2058-2063.杨娴,李勃,丁文,等.基于局部自适应色差阈值的彩色图像边缘检测[J].电子与信息学报,2012,34(9):2058-2063.
[18] Mullen K T.The contrast sensitivity of human color vision to red-green and blue-yellow chromatic gratings[J].The Journal of Physiology,1985,359(1):381-400.
[19] Wenzel K,Ladunga K,Samu K.Measurement of color defective and normal color vision subjects′ color and luminance contrast threshold functions on CRT[J].Periodica Polytechnica-Mechanical Engineering,2001,45(1):103-108.
[20] Yao J C.Measurements of human vision contrast sensitivity to opposite colors using a cathode ray tube display[J].Chinese Science Bullentin,2011,56(23):2425-2432.
[21] PENG Zhi-yong,CHANG Fa-liang,DONG Wen-hui.Vibe motion target detection algorithm based on lab color space[A].Advances in Image and Graphics Technologies (Communications in Computer and Information Science)[C].Berlin:Springer,2015,45-54.
[22] Goyette N,Jodoin P,Porikli F,et al.Changedetection.net:A new change detection benchmark dataset[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops[C].New York:IEEE,2012,1-8.
[2] AN Yun-feng,PAN Hai-peng.A new method for moving bubbles detection based on frame difference[J].Journal of Optoelectronics·Laser,2016,27(9):963-972.安云凤,潘海鹏.一种新的三帧差运动气泡检测方法[J].光电子·激光,2016,27(9):963-972.
[3] BAI Jun-qi.Infrared target detection method based on OTSU[J].Command Information System and Technology,2017,8(2):52-55.白俊奇.基于最大类间方差法的红外目标检测方法[J].指挥信息系统与技术,2017,8(2):52-55.
[4] Goyal K,Singhai J.Review of background subtraction methods using Gaussian mixture model for video surveillance systems[J].Artificial Intelligence Review,2017,(3):1-19.
[5] Liu W,Yu H,Yuan H,et al.Effective background modelling and subtraction approach for moving object detection[J].IET Computer Vision,2015,9(1):13-24.
[6] Barnich O,Droogenbroeck M.V.ViBe:A universal background subtraction algorithm for video sequences[J].IEEE Transactions on Image Processing,2011,20(6):1709-1724.
[7] YU Ye,CAO Ming-wei,YUE Feng.EVibe:An improved Vibe algorithm for detecting moving objects[J].Chinese Journal of Scientific Instrument,2014,35(4):924-931.余烨,曹明伟,岳峰.EVibe:一种改进的Vibe运动目标检测算法[J].仪器仪表学报,2014,35(4):924-931.
[8] CHEN Xing-ming,LIAO Juan,LI Bo,et al.Foreground detection based on modified ViBe in dynamic background[J].Optics and Precision Engineering,2014,22(9):2545-2552.陈星明,廖娟,李勃,等.动态背景下基于改进视觉背景提取的前景检测[J].光学精密工程,2014,22(9):2545-2552.
[9] MO Shao-wen,DENG Xin-pu,WANG Shuai,et al.Moving object detection algorithm based on improved visual background extractor[J].Acta Optica Sinica,2016,36(6):0615001.莫邵文,邓新蒲,王帅,等.基于改进视觉背景提取的运动目标检测算法[J].光学学报,2016,36(6):0615001.
[10] MIN Wen-dong,GUO Xiao-guang,HAN Qing.An improved ViBe algorithm and its application in traffic video processing[J].Optics and Precision Engineering,2017,25(3):806-811.闵卫东,郭晓光,韩清.改进的ViBe算法及其在交通视频处理中的应用[J].光学精密工程,2017,25(3):806-811.
[11] LIU Yong,XIA Dao-pin,QIU Qin-jun,et al.Research on visual background extractor to identify the vehicles based on edge similarity[J].International Journal of Wireless and Mobile Computing,2017,13(1):8-14.
[12] LUO Yuan,ZHOU Han-xing,TAN Qin,et al.Key frame extraction of surveillance video based on moving object detection and image similarity[J].Pattern Recognition and Image Analysis,2018,28(2):225-231.
[13] MIN Wei-dong,FAN Meng-dan,GUO Xiao-guang,et al.A new approach to track multiple vehicles with the combination of robust detection and two classifiers[J].IEEE Transactions on Intelligent Transportation Systems,2018,19(1):174-186.
[14] YANG Shi-yu,HAO Kuang-rong,DING Yong-sheng,et al.Improved visual background extractor with adaptive range change[J].Memetic Computing,2018,10(1):53-61.
[15] DING Ying,QIAN Feng,FAN Jing-tao,et al.Study on moving object detection algorithm based on different color space[J].Journal of Changchun University of Science and Technology (Natural Science Edition),2012,35(4):1-4.丁莹,钱锋,范静涛,等.基于不同颜色空间的运动目标检测算法分析[J].长春理工大学学报( 自然科学版),2012,35(4):1-4.
[16] WANG Ke,LU Chang-de,YUE Wan-de,et al.Color harmony system based on lab perceptual uniform color space[J].Journal of Northwestern Polytechnical University,2004,22(6):695-698.王可,陆长德,乐万德,等.基于Lab均匀色彩空间的色彩调和系统[J].西北工业大学学报,2004,22(6):695-698.
[17] YANG Xian,LI Bo,DING Wen,et al.Local adaptive perceptual color difference for color image edge detection[J].Journal of Electronics & Information Technology,2012,34(9):2058-2063.杨娴,李勃,丁文,等.基于局部自适应色差阈值的彩色图像边缘检测[J].电子与信息学报,2012,34(9):2058-2063.
[18] Mullen K T.The contrast sensitivity of human color vision to red-green and blue-yellow chromatic gratings[J].The Journal of Physiology,1985,359(1):381-400.
[19] Wenzel K,Ladunga K,Samu K.Measurement of color defective and normal color vision subjects′ color and luminance contrast threshold functions on CRT[J].Periodica Polytechnica-Mechanical Engineering,2001,45(1):103-108.
[20] Yao J C.Measurements of human vision contrast sensitivity to opposite colors using a cathode ray tube display[J].Chinese Science Bullentin,2011,56(23):2425-2432.
[21] PENG Zhi-yong,CHANG Fa-liang,DONG Wen-hui.Vibe motion target detection algorithm based on lab color space[A].Advances in Image and Graphics Technologies (Communications in Computer and Information Science)[C].Berlin:Springer,2015,45-54.
[22] Goyette N,Jodoin P,Porikli F,et al.Changedetection.net:A new change detection benchmark dataset[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops[C].New York:IEEE,2012,1-8.