基于运动检测与KCF算法的目标跟踪方法
|
摘要
近年来机器视觉技术广泛采用于各个领域,本文设计实现了一种基于运动检测与KCF算法的目标跟踪方法,该方法相对于传统单一的运动检测算法与目标跟踪算法在运动目标跟踪方面有很大优势。单一运动检测算法无法在目标停止运动后很好检测,而单一目标跟踪算法需要前期指定跟踪目标。本文设计的方法以运动检测作为触发条件,一旦检测到运动目标即开启目标跟踪器进行目标跟踪,因此可以对运动目标进行实时,高效,稳定的跟踪。该方法对于智能安防,视频监控技术有巨大应用潜力。
In recent years,machine vision technology has been widely used in various fields. In this paper,a target tracking method based on motion detection and KCF algorithm is designed and implemented. This method has a great advantage over the traditional single motion detection algorithm and target tracking algorithm in moving target tracking. The single motion detection algorithm can not detect well after the target stops moving,while the single target tracking algorithm needs to specify the tracking target in advance. This method uses motion detection as the trigger condition. Once the moving target is detected,the target tracking is started,so the moving target can be tracked in real time,efficiently and steadily. This method has great potential for intelligent security and video surveillance technology.
In recent years,machine vision technology has been widely used in various fields. In this paper,a target tracking method based on motion detection and KCF algorithm is designed and implemented. This method has a great advantage over the traditional single motion detection algorithm and target tracking algorithm in moving target tracking. The single motion detection algorithm can not detect well after the target stops moving,while the single target tracking algorithm needs to specify the tracking target in advance. This method uses motion detection as the trigger condition. Once the moving target is detected,the target tracking is started,so the moving target can be tracked in real time,efficiently and steadily. This method has great potential for intelligent security and video surveillance technology.
引文
[1]廖万辉,李琳.基于机器视觉的工业机器人定位系统[J].微计算机信息,2009,8:242-244.
[2]许海霞.机器视觉标定与目标检测跟踪方法及其应用研究[J].湖南大学学报,2011:230-232.
[3]龚爱平.基于嵌入式机器视觉的信息采集与处理技术研究[D].杭州:浙江大学,2013.
[4]周鑫,钱秋朦,叶永强,等.改进后的TLD视频目标跟踪方法[J].中国图象图形学报,2013,9:1115-1123.
[5]魏星.基于SVM的山体滑坡灾害图像识别方法[J].电子测量技术,2013,36(8):65-70.
[6]LIM T Y,RATNAM M M,KHALID M A.Automatic classification of weld defects using simulated data and an MLP neural network[J].Insight,2007,49(3):154-159.
[7]TSAI R.A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-theshelf TV cameras and lenses[J].IEEE Journal on Robotics and Automation,1987,3(4):323-344.
[8]INIGO R M,MCVEY E S,BERGER B J,et al.Machine vision applied to vehicle guidance[J].IEEE transactions on pattern analysis and machine intelligence,1984(6):820-826.
[9]MIRAPEIX J,GARCIA ALLENDE P B,COBO A,et a1.Real-time arc-welding defect detection and classification with principal component analysis and artificial neural networks[J].NDT&E International,2007,40(4):315-323.
[10]VINCENT P,LAROCHELLE H,LAJOIE I,et al.Stacked de noising auto encoders:learning useful representations in a deep network with a local de noising criterion[J].Journal of Machine Learning Research,2010,11(12):3371-3408.
[2]许海霞.机器视觉标定与目标检测跟踪方法及其应用研究[J].湖南大学学报,2011:230-232.
[3]龚爱平.基于嵌入式机器视觉的信息采集与处理技术研究[D].杭州:浙江大学,2013.
[4]周鑫,钱秋朦,叶永强,等.改进后的TLD视频目标跟踪方法[J].中国图象图形学报,2013,9:1115-1123.
[5]魏星.基于SVM的山体滑坡灾害图像识别方法[J].电子测量技术,2013,36(8):65-70.
[6]LIM T Y,RATNAM M M,KHALID M A.Automatic classification of weld defects using simulated data and an MLP neural network[J].Insight,2007,49(3):154-159.
[7]TSAI R.A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-theshelf TV cameras and lenses[J].IEEE Journal on Robotics and Automation,1987,3(4):323-344.
[8]INIGO R M,MCVEY E S,BERGER B J,et al.Machine vision applied to vehicle guidance[J].IEEE transactions on pattern analysis and machine intelligence,1984(6):820-826.
[9]MIRAPEIX J,GARCIA ALLENDE P B,COBO A,et a1.Real-time arc-welding defect detection and classification with principal component analysis and artificial neural networks[J].NDT&E International,2007,40(4):315-323.
[10]VINCENT P,LAROCHELLE H,LAJOIE I,et al.Stacked de noising auto encoders:learning useful representations in a deep network with a local de noising criterion[J].Journal of Machine Learning Research,2010,11(12):3371-3408.