A practical eye tracking algorithm is proposed in this paper. It tracks human face and eyes simultaneously to realize an integration of direct and indirect eye tracking. On one hand, human face is detected by Adaboost face detector, an eye localization approach is designed to localize the eye windows in the detected face area, for eye tracking, multi-visual features which are represented with statistic histograms are extracted from the eye window, and also be fused in particle filter (PF) tracking framework; On the other hand, CAMShift algorithm is also exploited to track the face area, and the eye detection approach is still used to detect reliable eye windows, so as to correct PF tracker. Experimental results show that the proposed algorithm is good at handling illumination changes, sudden face rotation, camera jitter, and partial occlusion. It also can recover from tracking failures in time, which improves the tracking performance in sustainable and robustness.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Leak detection and location play an important role in the management of a pipeline system. Some model-based methods, such as those based on the extended Kalman filter (EKF) or based on the strong tracking filter (STF), have been presented to solve this problem. But these methods need the nonlinear pipeline model to be linearized. Unfortunately, linearized transformations are only reliable if error propagation can be well approximated by a linear function, and this condition does not hold for a gas pipeline model. This will deteriorate the speed and accuracy of the detection and location. Particle filters are sequential Monte Carlo methods based on point mass (or ``particle'') representations of probability densities, which can be applied to estimate states in nonlinear and non-Gaussian systems without linearization. Parameter estimation methods are widely used in fault detection and diagnosis (FDD), and have been applied to pipeline leak detection and location. However, the standard particle filter algorithm is not applicable to time-varying parameter estimation. To solve this problem, artificial noise has to be added to the parameters, but its variance is difficult to determine. In this paper, we propose an adaptive particle filter algorithm, in which the variance of the artificial noise can be adjusted adaptively. This method is applied to leak detection and location of gas pipelines. Simulation results show that fast and accurate leak detection and location can be achieved using this improved particle filter.
W artykule opisane zostały algorytmy filtracji nieliniowej (rozszerzona filtracja Kalmana, bezśladowa filtracja Kalmana, bezśladowa filtracja Kalmana w wariancie rozszerzonym i filtracja cząstkowa) stosowana powszechnie do estymacji położenia. Dodatkowo zaprezentowane zostały rezultaty złożonych badań symulacyjnych porównujących jakość estymacji analizowanych rodzajów filtrów nieliniowych dla złożonej nieliniowości wektora stanu. Ocena jakości procesu filtracji została przeprowadzona w środowisku MATLAB. Przedstawione wyniki stanowią podstawę do projektowania dokładniejszych algorytmów estymacji położenia obiektu.
EN
In this paper several types of nonlinear filtrations (Extended Kalman Filtering, Unscended Kalman Filtering, Augmented Unscended Kalman Filtering and Particle Filtering) widely used to position estimation and their algirithms are described. Additionally complex simulations results, which are to compare abilities of analyzed nonlinear filtrates for different nonlinearities, are shown. The comparison of filtration quality was done in MATLAB environment. The presented results provide a basis for designing more accurate algorithms for object position estimation.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.