Search

Aiming to get better real time performance and effect of detection in dynamic scene of intelligent video surveillance system, non-parametric kernel density estimation (KDE) is used to model the background. And to solve the foreground detection is not precise enough, background subtraction method is fused to detect the foreground. And some modified work is done to suppress shadow and noise. Experiments show that the method proposed can get better real time performance and low noise detect result in intelligent video surveillance system.

In this paper, the global exponential stability is discussed for static neural networks with time varying delays. On the basis of the linear matrix inequalities (LMIs) technique, and Lyapunov functional method, we have obtained the main condition to ensure the global exponential stability of the equilibrium point for this system, which is dependent on the change rate of time varying delays. The proposed result is less restrictive than those given in the earlier literatures, easier to check in practice. Remarks are made with other previous works to show the superiority of the obtained results, and the simulation examples are used to demonstrate the effectiveness of our results.

Moving target tracking is the key part of intelligent visual surveillance system. Among the various tracking algorithms, the Beysian tracking algorithms and the kernel tracking algorithm are two algorithms that frequently used. The Beysian tracking algorithms mainly conclude Kalman filtering algorithm, extended Kalman filtering algorithm and particle filtering algorithm. Mean Shift is the most representative algorithm of the kernel target tracking. In this survey, the status and development of target tracking algorithms has been studied more extensively with providing a few examples of modified tracking algorithms. Then a comparison was presented based on the limitations and scope of applications. Finally, the paper showed further research prospects of moving target tracking are introduced.

Spectrum sensing performance of building indoor environment has been the focus of attention and research in low signal-to-noise ratio. In this paper, a primary users sensing approach to signal classification combining spectral correlation analysis and support vector machine (SVM) is introduced. Three spectral coherence characteristic parameters are chosen via spectral correlation analysis. By utilizing a nonlinear SVM, primary user signal has been detected. Simulations indicate that the overall success rate is above 90.2% when SNR is equal to-5dB and 80.1% in-15dB. Compared to the existing methods including the classifiers based on MME and ANN, the proposed approach is more effective in the case of low SNR and limited training numbers. The results show that the validity and superiority of the proposed algorithm in building indoor environment.

Energy self-supply is one of the most important problems, and there are many undesirable energy difficulties when carrier keeps working underwater, and. A method based on inertial pendulum has been presented, but the model of wave energy absorption efficiency which is a significant parameter is strongly nonlinear and difficult to be gained. A method based on radial basis function (RBF) neural network is adopted to model in time domain. To prevent the problem that structure and parameters of neural network are hard to be tuned, improved RBF network based on particle swarm optimization (PSO) is proposed. The results obtained by presented method and by simulation software ADAMS are compared, which validate the method used to be correct and make basis on future research.

In this paper, the global robust exponential stability is discussed for interval Cohen-Grossgerg neural network with time varying delays. On the basis of the linear matrix inequalities (LMIs) technique, and Lyapunov functional method combined with the Bellman inequality and Jensen inequality technique, we have obtained the main condition to ensure the global robust exponential stability of the equilibrium point for this system. The obtained stability criterion is dependent on the upper bound of time varying delays. The proposed result is less restrictive, and suitable of the cases of slow or fast time varying delays, easier to check in practice. Remarks are made with other previous works to show the superiority of the obtained result, and the simulation example is used to demonstrate the effectiveness of our result.

While the methods of gray image segmentation keeping innovation, researches on color image segmentation have a marked development, and the gray image segmentation provides research base for color image segmentation. At present, the problem of color image segmentation has become a fundamental one and had significant impact on both new pattern recognition algorithms and applications. In order to make the research in this field has a more comprehensive understanding, In this paper, several typical approaches are presented and discussed, and expected to seek an algorithm with better real-time and robustness. algorithm combining both SOFM and FCM clustering is selected to be better in robustness, and reduce the hassle of looking for a suitable feature space.

Because of high fire frequency and huge losses, the research of fire signal detection in the monitoring system is an important task in the fire-preventing field. The fire signal detection method based on vision can overcome the shortcomings that exist in some traditional methods i.e. it can surmount the large impact on environmental interference factors, such as temperature, photographic and smoke of environment. With many researcher’s results, it shows clearly that the error rate of flame recognition is low, and also the real-time ability and the anti-disturbance ability are very good.