High resolution cameras and multi camera systems are being used in areas of video surveillance like security of public places, traffic monitoring, and military and satellite imaging. This leads to a demand for computa...High resolution cameras and multi camera systems are being used in areas of video surveillance like security of public places, traffic monitoring, and military and satellite imaging. This leads to a demand for computational algorithms for real time processing of high resolution videos. Motion detection and background separation play a vital role in capturing the object of interest in surveillance videos, but as we move towards high resolution cameras, the time-complexity of the algorithm increases and thus fails to be a part of real time systems. Parallel architecture provides a surpass platform to work efficiently with complex algorithmic solutions. In this work, a method was proposed for identifying the moving objects perfectly in the videos using adaptive background making, motion detection and object estimation. The pre-processing part includes an adaptive block background making model and a dynamically adaptive thresholding technique to estimate the moving objects. The post processing includes a competent parallel connected component labelling algorithm to estimate perfectly the objects of interest. New parallel processing strategies are developed on each stage of the algorithm to reduce the time-complexity of the system. This algorithm has achieved a average speedup of 12.26 times for lower resolution video frames(320×240, 720×480, 1024×768) and 7.30 times for higher resolution video frames(1360×768, 1920×1080, 2560×1440) on GPU, which is superior to CPU processing. Also, this algorithm was tested by changing the number of threads in a thread block and the minimum execution time has been achieved for 16×16 thread block. And this algorithm was tested on a night sequence where the amount of light in the scene is very less and still the algorithm has given a significant speedup and accuracy in determining the object.展开更多
Mobile target tracking is a necessary function of some emerging application domains, such as virtual reality, smart home and intelligent healthcare. However, existing portable devices for target tracking are resource ...Mobile target tracking is a necessary function of some emerging application domains, such as virtual reality, smart home and intelligent healthcare. However, existing portable devices for target tracking are resource intensive and high-cost. Camera tracking is an effective location tracking way for those emerging applications which can reuse the existing ubiquitous video monitoring system. This paper proposes a dynamic community-based camera collaboration(D3C) framework for target location and tracking. The contributions of D3C mainly include that(1) nonlinear perspective projection model is selected as the camera sensing model and sequential Monte Carlo is employed to predict the target location;(2) a dynamic collaboration scheme is proposed, it is based on the local community-detection theory deriving from social network analysis. The performance of proposed approach is validated by both synthetic datasets and real-world application. The experiment results show that D3C meets the versatility, real-time and fault tolerance requirements of target tracking applications.展开更多
The multi-objective genetic algorithm(MOGA) is proposed to calibrate the non-linear camera model of a space manipulator to improve its locational accuracy. This algorithm can optimize the camera model by dynamic balan...The multi-objective genetic algorithm(MOGA) is proposed to calibrate the non-linear camera model of a space manipulator to improve its locational accuracy. This algorithm can optimize the camera model by dynamic balancing its model weight and multi-parametric distributions to the required accuracy. A novel measuring instrument of space manipulator is designed to orbital simulative motion and locational accuracy test. The camera system of space manipulator, calibrated by MOGA algorithm, is used to locational accuracy test in this measuring instrument. The experimental result shows that the absolute errors are [0.07, 1.75] mm for MOGA calibrating model, [2.88, 5.95] mm for MN method, and [1.19, 4.83] mm for LM method. Besides, the composite errors both of LM method and MN method are approximately seven times higher that of MOGA calibrating model. It is suggested that the MOGA calibrating model is superior both to LM method and MN method.展开更多
This paper proposes a self-position estimate algorithm for the multiple mobile robots; each robot uses two omnidirectional cameras and an accelerometer. In recent years, the Great East Japan Earthquake and large-scale...This paper proposes a self-position estimate algorithm for the multiple mobile robots; each robot uses two omnidirectional cameras and an accelerometer. In recent years, the Great East Japan Earthquake and large-scale disasters have occurred frequently in Japan. From this, development of the searching robot which supports the rescue team to perform a relief activity at a large-scale disaster is indispensable. Then, this research has developed the searching robot group system with two or more mobile robots. In this research, the searching robot equips with two omnidirectional cameras and an accelerometer. In order to perform distance measurement using two omnidirectional cameras, each parameter of an omnidirectional camera and the position and posture between two omnidirectional cameras have to be calibrated in advance. If there are few mobile robots, the calibration time of each omnidirectional camera does not pose a problem. However, if the calibration is separately performed when using two or more robots in a disaster site, etc., it will take huge calibration time. Then, this paper proposed the algorithm which estimates a mobile robot＇s position and the parameter of the position and posture between two omnidirectional cameras simultaneously. The algorithm proposed in this paper extended Nonlinear Transformation （NLT） Method. This paper conducted the simulation experiment to check the validity of the proposed algorithm. In some simulation experiments, one mobile robot moves and observes the circumference of another mobile robot which has stopped at a certain place. This paper verified whether the mobile robot can estimate position using the measurement value when the number of observation times becomes 10 times in n/18 of observation intervals. The result of the simulation shows the effectiveness of the algorithm.展开更多
In this paper, we propose a video searching system that utilizes face recognition as searching indexing feature. As the applications of video cameras have great increase in recent years, face recognition makes a perfe...In this paper, we propose a video searching system that utilizes face recognition as searching indexing feature. As the applications of video cameras have great increase in recent years, face recognition makes a perfect fit for searching targeted individuals within the vast amount of video data. However, the performance of such searching depends on the quality of face images recorded in the video signals. Since the surveillance video cameras record videos without fixed postures for the object, face occlusion is very common in everyday video. The proposed system builds a model for occluded faces using fuzzy principal component analysis(FPCA), and reconstructs the human faces with the available information. Experimental results show that the system has very high efficiency in processing the real life videos, and it is very robust to various kinds of face occlusions. Hence it can relieve people reviewers from the front of the monitors and greatly enhances the efficiency as well. The proposed system has been installed and applied in various environments and has already demonstrated its power by helping solving real cases.展开更多
文摘High resolution cameras and multi camera systems are being used in areas of video surveillance like security of public places, traffic monitoring, and military and satellite imaging. This leads to a demand for computational algorithms for real time processing of high resolution videos. Motion detection and background separation play a vital role in capturing the object of interest in surveillance videos, but as we move towards high resolution cameras, the time-complexity of the algorithm increases and thus fails to be a part of real time systems. Parallel architecture provides a surpass platform to work efficiently with complex algorithmic solutions. In this work, a method was proposed for identifying the moving objects perfectly in the videos using adaptive background making, motion detection and object estimation. The pre-processing part includes an adaptive block background making model and a dynamically adaptive thresholding technique to estimate the moving objects. The post processing includes a competent parallel connected component labelling algorithm to estimate perfectly the objects of interest. New parallel processing strategies are developed on each stage of the algorithm to reduce the time-complexity of the system. This algorithm has achieved a average speedup of 12.26 times for lower resolution video frames(320×240, 720×480, 1024×768) and 7.30 times for higher resolution video frames(1360×768, 1920×1080, 2560×1440) on GPU, which is superior to CPU processing. Also, this algorithm was tested by changing the number of threads in a thread block and the minimum execution time has been achieved for 16×16 thread block. And this algorithm was tested on a night sequence where the amount of light in the scene is very less and still the algorithm has given a significant speedup and accuracy in determining the object.
基金supported by National Natural Science Foundation of China (Grant No. 61501048) National High-tech R&D Program of China (863 Program) (Grant No. 2013AA102301)+1 种基金The Fundamental Research Funds for the Central Universities (Grant No. 2017RC12) China Postdoctoral Science Foundation funded project (Grant No.2016T90067, 2015M570060)
文摘Mobile target tracking is a necessary function of some emerging application domains, such as virtual reality, smart home and intelligent healthcare. However, existing portable devices for target tracking are resource intensive and high-cost. Camera tracking is an effective location tracking way for those emerging applications which can reuse the existing ubiquitous video monitoring system. This paper proposes a dynamic community-based camera collaboration(D3C) framework for target location and tracking. The contributions of D3C mainly include that(1) nonlinear perspective projection model is selected as the camera sensing model and sequential Monte Carlo is employed to predict the target location;(2) a dynamic collaboration scheme is proposed, it is based on the local community-detection theory deriving from social network analysis. The performance of proposed approach is validated by both synthetic datasets and real-world application. The experiment results show that D3C meets the versatility, real-time and fault tolerance requirements of target tracking applications.
基金Project(J132012C001)supported by Technological Foundation of ChinaProject(2011YQ04013606)supported by National Major Scientific Instrument & Equipment Developing Projects,China
文摘The multi-objective genetic algorithm(MOGA) is proposed to calibrate the non-linear camera model of a space manipulator to improve its locational accuracy. This algorithm can optimize the camera model by dynamic balancing its model weight and multi-parametric distributions to the required accuracy. A novel measuring instrument of space manipulator is designed to orbital simulative motion and locational accuracy test. The camera system of space manipulator, calibrated by MOGA algorithm, is used to locational accuracy test in this measuring instrument. The experimental result shows that the absolute errors are [0.07, 1.75] mm for MOGA calibrating model, [2.88, 5.95] mm for MN method, and [1.19, 4.83] mm for LM method. Besides, the composite errors both of LM method and MN method are approximately seven times higher that of MOGA calibrating model. It is suggested that the MOGA calibrating model is superior both to LM method and MN method.
文摘This paper proposes a self-position estimate algorithm for the multiple mobile robots; each robot uses two omnidirectional cameras and an accelerometer. In recent years, the Great East Japan Earthquake and large-scale disasters have occurred frequently in Japan. From this, development of the searching robot which supports the rescue team to perform a relief activity at a large-scale disaster is indispensable. Then, this research has developed the searching robot group system with two or more mobile robots. In this research, the searching robot equips with two omnidirectional cameras and an accelerometer. In order to perform distance measurement using two omnidirectional cameras, each parameter of an omnidirectional camera and the position and posture between two omnidirectional cameras have to be calibrated in advance. If there are few mobile robots, the calibration time of each omnidirectional camera does not pose a problem. However, if the calibration is separately performed when using two or more robots in a disaster site, etc., it will take huge calibration time. Then, this paper proposed the algorithm which estimates a mobile robot＇s position and the parameter of the position and posture between two omnidirectional cameras simultaneously. The algorithm proposed in this paper extended Nonlinear Transformation （NLT） Method. This paper conducted the simulation experiment to check the validity of the proposed algorithm. In some simulation experiments, one mobile robot moves and observes the circumference of another mobile robot which has stopped at a certain place. This paper verified whether the mobile robot can estimate position using the measurement value when the number of observation times becomes 10 times in n/18 of observation intervals. The result of the simulation shows the effectiveness of the algorithm.
基金supported by the National Natural Science Foundation of China(No.61502256)
文摘In this paper, we propose a video searching system that utilizes face recognition as searching indexing feature. As the applications of video cameras have great increase in recent years, face recognition makes a perfect fit for searching targeted individuals within the vast amount of video data. However, the performance of such searching depends on the quality of face images recorded in the video signals. Since the surveillance video cameras record videos without fixed postures for the object, face occlusion is very common in everyday video. The proposed system builds a model for occluded faces using fuzzy principal component analysis(FPCA), and reconstructs the human faces with the available information. Experimental results show that the system has very high efficiency in processing the real life videos, and it is very robust to various kinds of face occlusions. Hence it can relieve people reviewers from the front of the monitors and greatly enhances the efficiency as well. The proposed system has been installed and applied in various environments and has already demonstrated its power by helping solving real cases.
