目的:单摄像机视觉跟踪过程中,常发生目标被遮挡或背景复杂的情况,此时容

目的:单摄像机视觉跟踪过程中,常发生目标被遮挡或背景复杂的情况,此时容易跟丢目标,为了提高跟踪的准确性.方法:本文从目标表现和背景的不确定性入手,以协方差特征对目标表现以及背景进行建模,同时将序贯拟蒙特卡洛滤波作为跟踪框架应用于视觉目标跟踪,优化采样粒子的分布,在估计粒子的权重时,不仅考虑目标的真实状态和可能状态的相似性,还考虑了目标可能的状态和背景的差异.:结果:将本文提出的算法与粒子滤波,均值漂移,基于协方差概率跟踪算法进行比较,通过Matlab2010编程平台,比较了几种算法的处理速度以及跟踪误差,试验结果表明本文提出的算法每秒处理速度为60帧/秒,优于上述三种跟踪算法平均误差值也高于另外三种算法..结论:本文提出算法在目标存在遮挡和背景较为复杂时,能够保证对目标进行准确,连续的跟踪.:

結果 (英文) 1:

Purpose: single camera visual track process in the, often occurred target was block or background complex of situation, at this time easy with lost target, to improve track of accuracy. method: this from target performance and background of uncertainty start, to Association variance features on target performance and background for built die, while will sequence sequential intends Monte Carlo filtering as track framework application Yu Visual target track, optimization sampling particle of distribution, in estimated particle of weights Shi, not only consider target of real state and may state of similar sexual, Also takes into account the differences of status and background of the target may be.: results: will the proposed algorithm and particle filter, mean shift,Comparison of probabilistic tracking algorithm based on covariance, Matlab2010 programming platform, comparison of several algorithms for processing speed, as well as tracking error, test results show that the proposed algorithm is processed per second speed of 60 frames per second, better than the three tracking average error value is higher than the other three algorithms ... Conclusion this paper when there is occlusion and background are more complex algorithm in target, to ensure that targets for accurate, continuous track.

結果 (英文) 2:

Case of a single camera visual tracking process often occurs goal is blocked or complex background, this time with the lost easy target, in order to improve the accuracy of tracking methods:: The purpose of this paper, the performance from the target and the background of uncertainty start to co- variance performance characteristics of the target and the background modeling, while the sequential Monte Carlo filter proposed framework as applied to visual target tracking to track and optimize the distribution of the sample particles, heavy particles in the right estimate, not only to consider the true state of the target and may similarity of the state, but also consider the differences between the target and the background of possible states: Results: The proposed algorithm and particle filter, mean shift tracking algorithm based on the probability of covariance comparing programming platform via Matlab2010 compare several the processing speed of the algorithm and the tracking error, the test results show that the proposed algorithm per second processing speed of 60 frames / sec, better than the average of the three tracking algorithm error value is also higher than the other three algorithms .. Conclusion: This paper presents algorithms target occlusion exists when the background is complex, the target to ensure accurate and continuous tracking:

結果 (英文) 3:

Objective: single camera tracking process, the target is occluded or complex background often occurs, the easy to lose the goal, in order to improve the tracking accuracy. Methods: This article from the target and the background of uncertainty, the covariance features of target and background modeling performance, at the same time sequential Quasi Monte Carlo filtering as the tracking framework is applied to the visual target tracking, the optimization of distribution of particle, in the estimation of the weight of the particle, not only consider the similarity of the true state of the target and the possible states, considering the potential target and background difference. Results: the algorithm and particle filter is proposed in this paper, mean shift,Covariance tracking algorithm based on probability is compared, through Matlab2010 programming platform, the processing speed of several algorithms and the tracking error, test results show that the algorithm proposed in this paper for the second the speed of 60 frames / second, better than the three average error tracking algorithm is also higher than the other three algorithms. Conclusion: the proposed algorithm of the occlusion and background on the target complex, able to guarantee the accurate, continuous tracking: